Operation optimization method, operation optimization program, and storage medium

Abstract

This operation optimization method comprises: a step for acquiring a data set which includes a plurality of parameters and a plurality of pieces of factor data and which indicates an operation result of a target system; a step for using at least some of the plurality of pieces of factor data to set a group of numerical values for an index to be optimized; a step for extracting, from the numerical value group for the index, a first numerical value group included in a first period and a second numerical value group included in a second period; a step for using the variation in the first numerical value group to determine a target value which is a variable for optimizing the second numerical value group; a step for extracting, from the first numerical value group, a third numerical value group which is a group of numerical values that have achieved the target value; a step for analyzing the deviation between a data set corresponding to the third numerical value group and a data set corresponding to the second numerical value group; and a step for outputting result data created using the data set corresponding to the third numerical value group, on the basis of the deviation.

Description

Driving optimization method, driving optimization program, and storage medium

[0001] The present disclosure relates to a driving optimization method, a driving optimization program, and a storage medium.

[0002] There are known techniques for simulating the operation of a plant, etc. Patent Document 1 describes a technique for proposing control parameters using a plant model that simulates the operation of the entire plant and a structural model that calculates the temperature distribution of a rotating machine, etc. Patent Document 2 describes a technique for training a neural network using learning input data based on the time response, etc., of a control system model and control parameters as training data.

[0003] JP 2021-135871 A JP 3-233702 A

[0004] Simulations using conventional physical models tend to be large-scale and complex. Such physical models impose a heavy simulation load and increase computational costs. As a result, the scale of facilities or businesses that can build physical models is limited.

[0005] The present disclosure describes a technique that can reduce the computational costs of data that optimizes the operation of a target system.

[0006] A method for optimizing operation according to one aspect of the present disclosure is executed by a computer. The method includes: an acquisition step of acquiring a dataset indicating the operating performance of a target system, the dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters; an index setting step of setting a set of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the set of numerical values ​​of the indexes, a first set of numerical values ​​included in a first period that is a predetermined period and a second set of numerical values ​​included in a second period that is a portion of the first period; a target determination step of determining, using a variance in the first set of numerical values, a target value that is a variable for optimizing the second set of numerical values; a second extraction step of extracting, from the first set of numerical values, a third set of numerical values ​​that is a set of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third set of numerical values ​​and a dataset corresponding to the second set of numerical values; and an output step of outputting result data created using the dataset corresponding to the third set of numerical values ​​based on the deviation.

[0007] According to the present disclosure, it is possible to provide a technology that can reduce the calculation cost of data for optimizing the operation of a target system.

[0008] FIG. 1 is a block diagram showing an example of the entire configuration including a driving optimization system. FIG. 2 is a diagram showing an example of data used in the driving optimization system. FIG. 3 is a diagram showing an example of the variance of a first group of values ​​and an example of a second group of values. FIG. 4 is a diagram showing an example of the extraction result of a third group of values. FIG. 5 is a diagram showing an example of result data. FIG. 6 is a flowchart showing an example of the operation of a driving optimization device. FIG. 7 is a diagram showing an example of a hardware configuration related to the driving optimization system.

[0009] A method for optimizing operation according to one aspect of the present disclosure is executed by a computer. The method includes: an acquisition step of acquiring a dataset indicating the operating performance of a target system, the dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters; an index setting step of setting a set of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the set of numerical values ​​of the indexes, a first set of numerical values ​​included in a first period that is a predetermined period and a second set of numerical values ​​included in a second period that is a portion of the first period; a target determination step of determining, using a variance in the first set of numerical values, a target value that is a variable for optimizing the second set of numerical values; a second extraction step of extracting, from the first set of numerical values, a third set of numerical values ​​that is a set of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third set of numerical values ​​and a dataset corresponding to the second set of numerical values; and an output step of outputting result data created using the dataset corresponding to the third set of numerical values ​​based on the deviation.

[0010] An operation optimization program according to one aspect of the present disclosure causes a computer to execute the following steps: an acquisition step for acquiring a dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters, and indicating the operating performance of a target system; an index setting step for setting a group of numerical values ​​of indicators to be optimized using at least a portion of the plurality of factor data; a first extraction step for extracting, from the group of numerical values ​​of the indicators, a first group of numerical values ​​included in a first period, which is a predetermined period, and a second group of numerical values ​​included in a second period, which is a portion of the first period; a target determination step for determining a target value, which is a variable for optimizing the second group of numerical values, using the variance of the first group of numerical values; a second extraction step for extracting, from the first group of numerical values, a third group of numerical values ​​that is a group of numerical values ​​that achieves the target value; an analysis step for analyzing the deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step for outputting result data created using a dataset corresponding to the third group of numerical values ​​based on the deviation.

[0011] A storage medium according to one aspect of the present disclosure is a computer-readable storage medium. The storage medium has recorded thereon an operation optimization program including: an acquisition step of acquiring a dataset indicating the operating performance of a target system, the dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters; an index setting step of setting a set of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the set of numerical values ​​of the indexes, a first set of numerical values ​​included in a first period that is a predetermined time period and a second set of numerical values ​​included in a second period that is a portion of the first period; a target determination step of determining, using a variance in the first set of numerical values, a target value that is a variable for optimizing the second set of numerical values; a second extraction step of extracting, from the first set of numerical values, a third set of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third set of numerical values ​​and a dataset corresponding to the second set of numerical values; and an output step of outputting result data created using the dataset corresponding to the third set of numerical values ​​based on the deviation.

[0012] In a driving optimization method, a driving optimization program, and a storage medium according to one aspect of the present disclosure, a set of values ​​of an index to be optimized is set using at least a portion of multiple factor data. A first set of values ​​for a first period and a second set of values ​​for a second period are extracted from the set of values ​​of the index. A target value, which is a variable for optimizing the second set of values, is determined based on the variance of the first set of values. A third set of values ​​that achieves the target value is then extracted from the first set of values. Based on the deviation between a data set corresponding to the third set of values ​​and a data set corresponding to the second set of values, result data created using a data set corresponding to the third set of values ​​is output. That is, from the first set of values ​​of the entire set, a third set of values ​​of a subset that achieves the target value (has good efficiency) is extracted, and data causing the deviation between the third set of values ​​and the second set of values ​​is analyzed from a data set including multiple parameters and multiple factor data. The result data based on the deviation can also be considered data aiming to be champion data when efficiency is good. That is, the result data is data that can optimize the operation of the target system. The process of the present disclosure eliminates the need to create models such as physical models, thereby reducing computational costs. As a result, the computational costs for data that optimizes the operation of the target system can be reduced.

[0013] The first extraction step may extract a first set of values ​​from the entire first period, and extract a second set of values ​​from the second period including the latest time information. In this case, a sufficient number of data points for the first set of values ​​in the entire set is ensured, and a second set of values ​​based on the latest operation of the target system is extracted. This makes it possible to output data that optimizes the latest operation of the target system.

[0014] The first extraction step may include narrowing down at least some of the parameters and factor data included in the first time period using extraction conditions based on at least some of the parameters and factor data included in the second time period, thereby extracting a first set of numerical values. In this case, the first set of numerical values ​​is narrowed down based on at least some of the parameters and factor data included in the second time period. That is, the first set of numerical values ​​is extracted when the data set is at least partially similar to the data set of the second time period. By using such data, the accuracy of optimizing the operation of the target system can be improved.

[0015] The first extraction step may include narrowing down at least some of the parameters and factor data included in the first period, using an extraction condition that the parameters and factor data included in the first period are normal, to extract a first group of numerical values. In this case, the first group of numerical values ​​is narrowed down based on at least some of the normal parameters and factor data. That is, the first group of numerical values ​​is extracted based on normal data. Using normal data can improve the accuracy of optimizing the operation of the target system.

[0016] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the description of the drawings, the same elements are designated by the same reference numerals, and duplicated description will be omitted.

[0017] 1 is a block diagram showing an example of the overall configuration including a driving optimization system 1. For example, the driving optimization system 1 is configured to include a driving optimization device 10. The driving optimization device 10 is connected to a target system 20, an external system 30, and a terminal 40 so as to be able to communicate with them.

[0018] The target system 20 is, for example, a plant, but is not limited to this. In the present disclosure, an example will be described in which the target system 20 is a coal-fired boiler plant. The target system 20 includes, for example, multiple devices (e.g., a boiler, a mill, a soot blower, etc.), a control device that controls the multiple devices, and multiple sensors that measure the status of the multiple devices. The target system 20 transmits an internal dataset, which is a collection of data that can be acquired within the target system 20, to the operation optimization device 10. The data included in the internal dataset is not limited. For example, the internal dataset may include values ​​manipulated by an operator or the like to operate the target system 20, and values ​​obtained as a result of operating the target system 20. In one example, the internal dataset may include time information, various parameters, material information, fuel information (e.g., fuel type, moisture content, quality, and input amount), measurement values ​​of multiple sensors, yield, power consumption, etc.

[0019] The external system 30 is, for example, but is not limited to, a weather forecast system. The external system 30 transmits an external dataset, which is a collection of data that can be acquired outside the target system 20 (e.g., the external system 30), to the driving optimization device 10. The data included in the external dataset is not limited. For example, the external dataset may include values ​​that may be noise regarding the operation of the target system 20. In one example, the external dataset may include time information, outside temperature, humidity, weather information, etc.

[0020] The terminal 40 is one or more computers used by a user of the driving optimization system 1. The type of the terminal 40 is not limited. For example, the terminal 40 may be a personal computer. The terminal 40 may also be a high-function mobile phone (smartphone), a tablet terminal, a wearable terminal, or the like.

[0021] The driving optimization device 10 is a device that outputs information for optimizing the driving of a target system 20. The driving optimization device 10 includes a database 12. The driving optimization device 10 includes, as functional elements, an acquisition unit 11, an index setting unit 13, a first extraction unit 14, a target determination unit 15, a second extraction unit 16, an analysis unit 17, and an output unit 18.

[0022] The database 12 is a non-transitory storage medium or storage device that stores various information used by the driving optimization system 1. The database 12 may be constructed as a single database or a collection of multiple databases. The location of the database 12 is not limited. For example, the database 12 may be provided in a computer system separate from the driving optimization system 1.

[0023] The acquisition unit 11 acquires a dataset indicating the operating performance of the target system 20, including multiple parameters and multiple factor data other than the multiple parameters. The multiple parameters are values ​​manipulated by an operator or the like to operate the target system 20. The multiple parameters may also include values ​​calculated based on measurement values ​​of multiple sensors. The factor data is data obtained as a result of operating the target system 20, values ​​that may become noise regarding the operation of the target system 20, etc. The factor data may include values ​​that are controllable and uncontrollable during the operation of the target system 20. For example, the acquisition unit 11 acquires an internal dataset from the target system 20. The acquisition unit 11 acquires an external dataset from the external system 30. The acquisition unit 11 creates a dataset including multiple parameters and multiple factor data using the internal dataset and the external dataset. For example, the acquisition unit 11 may create a dataset by combining the internal dataset and the external dataset based on time information. The acquisition period and acquisition interval of the dataset are not limited. In one example, the acquisition unit 11 may acquire a dataset including multiple parameters and multiple factor data every 10 minutes for one year, or every minute for six months. In another example, the acquisition unit 11 may acquire a dataset including multiple parameters and multiple factor data along an arbitrary time series, without being limited to a fixed interval. The acquisition unit 11 stores the dataset in the database 12.

[0024] The index setting unit 13 sets a set of numerical values ​​of indexes to be optimized using at least a part of the plurality of factor data. For example, the index setting unit 13 sets a set of numerical values ​​of indexes to be optimized for each piece of time information using at least a part of the plurality of factor data. The indexes to be optimized may be information indicating the operating efficiency of the target system 20. Examples of indexes to be optimized include heat loss, power generation amount, CO ２ Examples of the index to be optimized include, but are not limited to, emissions, steam volume, etc. The index setting unit 13 may receive input from a user to select an index to be optimized. The type of factor data used for the calculation may vary depending on the index to be optimized. In one example, the index setting unit 13 may calculate heat loss using measured values ​​of multiple sensors, etc. Heat loss can be expressed as, for example, L1 [kg / kJ], etc. In another example, the index setting unit 13 may calculate CO ２ The indicator setting unit 13 may store the calculated numerical values ​​of the indicators in the database 12.

[0025] The index setting unit 13 may set specific factor data of the dataset as the index to be optimized. In this case, the index setting unit 13 sets a set of numerical values ​​of the specific factor data as the set of numerical values ​​of the index to be optimized. In one example, the index setting unit 13 may set the yield as the index to be optimized. The index setting unit 13 may set a set of numerical values ​​of the yield as the set of numerical values ​​of the index to be optimized. In other words, the index setting unit 13 does not need to calculate a new set of numerical values ​​depending on the index to be optimized.

[0026] The first extraction unit 14 extracts, from the set of index values, a first set of values ​​included in a first period, which is a predetermined period, and a second set of values ​​included in a second period, which is a portion of the first period. For example, the first extraction unit 14 may extract the first set of values ​​by setting the first period to the entire period (e.g., one year or six months). The first extraction unit 14 may extract the second set of values ​​by setting the second period to a period including the latest time information (e.g., the most recent week).

[0027] The first extraction unit 14 may use extraction conditions based on at least some of the parameters and factor data included in the second period to narrow down at least some of the parameters and factor data included in the first period and extract a first set of values. For example, the first extraction unit 14 may use the values ​​of a specific parameter or specific factor data included in the second period as a reference to extract a first set of values ​​when the values ​​of the specific parameter or specific factor data included in the first period are within a predetermined range. In other words, the first extraction unit 14 may narrow down data when the operating performance data for the first period and the operating performance data for the second period are under similar conditions. Here, "similar" may mean that the first set of values ​​follows a normal distribution due to a specific condition being within a certain range. The first set of values ​​may also be data that follows a continuous distribution or a discrete distribution. For example, when the mill inlet temperature at the latest time information is x°C, the first extraction unit 14 may extract a first set of values ​​when the mill inlet temperature included in the entire period is x±10°C. In other words, the first extraction unit 14 extracts past data when the mill inlet temperature was similar to the most recent mill inlet temperature.

[0028] The first extractor 14 may extract a first set of numerical values ​​by narrowing down at least some of the parameters and factor data included in the first period, using an extraction condition that the parameters and factor data included in the first period are normal. In other words, the first extractor 14 may extract a first set of numerical values ​​excluding abnormal data. In one example, the first extractor 14 may determine an abnormality when at least some of the parameters and factor data cannot be acquired. In another example, the first extractor 14 may determine an abnormality when at least some of the parameters and factor data include outliers. In yet another example, the first extractor 14 may determine an abnormality when information indicating an error (such as a flag) is associated with the dataset. In yet another example, the first extractor 14 may determine an abnormality for a certain period defined by a user as an abnormality.

[0029] The target determination unit 15 uses the variation in the first set of values ​​to determine a target value, which is a variable that optimizes the second set of values. For example, the target determination unit 15 calculates the average value and standard deviation of the first set of values. The target determination unit 15 determines a value within the standard deviation of the first set of values ​​as the target value. In other words, the target determination unit 15 determines a target value that is expected to improve the operating efficiency of the second set of values ​​within the range of the proven first set of values. That is, the target determination unit 15 determines the target value based on the extent to which improvement of the second set of values ​​can be achieved. In one example, the target determination unit 15 may determine 1σ in the standard deviation as the target.

[0030] The target determination unit 15 may graph the average value and standard deviation of the first group of values ​​and the second group of values, and display them on the terminal 40. The target determination unit 15 may accept input of a target value from a user. In one example, the target determination unit 15 may graph the range of ±3σ for the variation of the first group of values. The target determination unit 15 may accept input from a user who aims for 1σ.

[0031] The second extraction unit 16 extracts a third group of values ​​from the first group of values, which is a group of values ​​that achieves the target value. For example, the second extraction unit 16 extracts a third group of values ​​that is further away from the second group of values ​​by at least the target value. That is, the second extraction unit 16 extracts, from the first group of values, a group of values ​​that resulted in better operating efficiency than the second group of values, as the third group of values.

[0032] The analysis unit 17 analyzes the deviation between the dataset corresponding to the third group of values ​​and the dataset corresponding to the second group of values. The "dataset corresponding to the group of values" refers to a dataset including one or more factor data used to set the group of values. The "dataset corresponding to the group of values" can also be said to be a dataset obtained by reverse lookup from the group of values. For example, the analysis unit 17 compares a dataset corresponding to a time when driving efficiency was good (third group of values) with the dataset of the second group of values ​​to analyze which data deviates. The deviating data can be said to be data that can contribute to improving the driving efficiency of the second group of values. The target of the analysis may be at least a portion, or all, of the multiple parameters and multiple factor data.

[0033] The analysis unit 17 may perform the analysis using an abnormality diagnosis method assuming that the third group of values ​​represents a normal state and the second group of values ​​represents an abnormal state. Examples of the analysis method include, but are not limited to, the Mahalanobis-Taguchi Method (MT), principal component analysis, and clustering. For example, the analysis unit 17 uses the MT method to analyze at least a portion of the multiple parameters and multiple factor data in order of contribution. The contribution indicates a breakdown of the abnormality score in the MT method. It can be said that a higher contribution is more likely to contribute to improving operating efficiency. In one example, the analysis unit 17 may analyze that a deviation in the operation interval of a specific soot blower contributes most to reducing heat loss (improving fuel efficiency).

[0034] The output unit 18 outputs result data created using the data set corresponding to the third group of values ​​based on the deviation. For example, the output unit 18 may output the average value and standard deviation calculated for the deviated data using the data set of the third group of values ​​as result data. In one example, the output unit 18 may output the average value and standard deviation for the operation interval of a specific soot blower as result data. In another example, the output unit 18 may output the calculated average value and standard deviation for the amount of fuel input as result data. The output unit 18 may transmit the result data to the terminal 40 or another device. The output unit 18 may display the result data on a display device provided in the operation optimization device 10 or another device.

[0035] FIG. 2 is a diagram showing an example of data used in the operation optimization system 1. The database 12 stores multiple parameters P and multiple factor data F in association with time information T. The index setting unit 13 sets a set of numerical values ​​of an index K to be optimized using at least a portion of the multiple factor data F. FIG. 2 shows heat loss as an example of the index K. As shown in FIG. 2, the index K to be optimized is set for each piece of time information T. In one example, when the time information T is 2022 / 8 / 5 0:00, the heat loss is set as L1 [kJ / kg]. Furthermore, in FIG. 2, it can be said that a data set including multiple parameters P and multiple factor data F is associated with the numerical value of the index K to be optimized for each piece of time information T. The index setting unit 13 may store the calculated set of numerical values ​​of the index in the database 12.

[0036] FIG. 3 is a diagram showing an example of the variation of the first group of values ​​and the second group of values. In the graph G1 shown in FIG. 3, the vertical axis represents heat loss [kJ / kg], and the horizontal axis represents time. Graph G1 displays an average value AV1 of the first group of values, a standard deviation SD1 of the first group of values, and a second group of values ​​DX1. The standard deviation SD1 indicates a range of ±3σ. In one example, the target determination unit 15 may determine 1σ as the target value. That is, the target determination unit 15 may determine a value 1σ below the average value (a value with improved low heat loss) as the target value.

[0037] FIG. 4 is a diagram showing an example of the extraction results of the third group of values. In the graph G2 shown in FIG. 4, the vertical axis represents heat loss [kJ / kg], and the horizontal axis represents time. Graph G2 displays the average value AV2 of the third group of values, the standard deviation SD2 of the third group of values, and the second group of values ​​DX1. The standard deviation SD2 indicates a range of ±3σ. In the example of FIG. 4, the average value AV2 and standard deviation SD2 of the third group of values ​​that are 1σ or more away from the second group of values ​​DX1 are displayed. That is, graph G2 is an example of the results showing the variation of the third group of values ​​extracted, where the third group of values ​​has lower heat loss (good fuel efficiency) than the second group of values ​​DX1 by 1σ or more.

[0038] In graph G2, the third group of values ​​that is less than 1σ based on the second group of values ​​DX1 is not extracted, and the average value AV2 and standard deviation SD2 of the third group of values ​​in this case are not shown in graph G2.

[0039] FIG. 5 is a diagram showing an example of result data. In graph G3 shown in FIG. 5, the vertical axis represents the elapsed time for the operation of the equipment, and the horizontal axis represents time. Graph G3 uses plot PL to show the elapsed time during which a specific soot blower corresponding to the second group of values ​​operated. Graph G3 also shows the average value AV3 and standard deviation SD3 of the elapsed time during which a specific soot blower corresponding to the third group of values ​​operated. In graph G3, the average value AV3 is 145 minutes, and the standard deviation SD3 is approximately ±100 minutes. In other words, when the efficiency was high (third group of values), the specific soot blower operated for approximately 145 minutes ±100 minutes. In contrast, plot PL for the second group of values ​​rises to the right. In other words, the period of the rise can be said to be the time during which the specific soot blower corresponding to the second group of values ​​did not operate again. Graph G can be said to be a graph showing the deviation between the operation of a specific soot blower corresponding to good operating efficiency (third numerical group) and the operation of a specific soot blower corresponding to the second numerical group.

[0040] [Driving optimization method] An example of an operation method (driving optimization method) by the driving optimization device 10 will be described with reference to Fig. 6. Fig. 6 is a flowchart showing an example of the operation of the driving optimization device 10.

[0041] In step S1 (acquisition step), the acquisition unit 11 acquires a dataset that includes a plurality of parameters and a plurality of factor data other than the plurality of parameters and indicates the operating performance of the target system 20. For example, the acquisition unit 11 acquires an internal dataset from the target system 20. The acquisition unit 11 acquires an external dataset from the external system 30. The acquisition unit 11 creates a dataset that includes a plurality of parameters and a plurality of factor data using the internal dataset and the external dataset. The acquisition unit 11 stores the dataset in the database 12. In one example, the acquisition unit 11 may store in the database 12 a dataset that includes a plurality of parameters P and a plurality of factor data F as shown in FIG. 2 .

[0042] In step S2 (index setting step), the index setting unit 13 sets a set of numerical values ​​of the indexes to be optimized using at least a part of the multiple factor data. The index setting unit 13 may receive input from a user selecting the indexes to be optimized. The type of factor data used for the calculation may differ depending on the indexes to be optimized. The index setting unit 13 may store the calculated set of numerical values ​​of the indexes in the database 12.

[0043] The index setting unit 13 may set specific factor data of the dataset as the index to be optimized. In this case, the index setting unit 13 sets a group of numerical values ​​of the specific factor data as a group of numerical values ​​of the index to be optimized.

[0044] In step S3 (first extraction step), the first extraction unit 14 extracts, from the group of index values, a first group of values ​​included in a first period, which is a predetermined period, and a second group of values ​​included in a second period, which is a portion of the first period.

[0045] The first extraction unit 14 may use extraction conditions based on at least a portion of the multiple parameters and multiple factor data included in the second period to narrow down at least a portion of the multiple parameters and multiple factor data included in the first period, and extract a first group of numerical values.

[0046] The first extracting unit 14 may extract the first group of numerical values ​​by narrowing down at least some of the parameters and factor data included in the first period, using an extraction condition that the parameters and factor data included in the first period are normal. In other words, the first extracting unit 14 may extract the first group of numerical values ​​excluding abnormal data.

[0047] In step S4 (target determination step), the target determination unit 15 determines a target value, which is a variable that optimizes the second group of values, using the variance of the first group of values. For example, the target determination unit 15 calculates the average value and standard deviation of the first group of values. The target determination unit 15 determines a value within the standard deviation of the first group of values ​​as the target value.

[0048] The goal determination unit 15 may graph the average value and standard deviation of the first group of values ​​and the second group of values, and display the graph on the terminal 40. The goal determination unit 15 may accept an input of a target value from a user. In one example, the goal determination unit 15 may display the graph shown in FIG. 3 on the terminal 40.

[0049] In step S5 (second extraction step), the second extraction unit 16 extracts a third group of values ​​from the first group of values, which is a group of values ​​that achieves the target value. For example, the second extraction unit 16 extracts a third group of values ​​that is further away from the second group of values ​​by at least the target value. That is, the second extraction unit 16 extracts, from the first group of values, a group of values ​​that results in better operating efficiency than the second group of values, as the third group of values. In one example, the second extraction unit 16 may extract a third group of values ​​as shown in FIG. 4.

[0050] In step S6 (analysis step), the analysis unit 17 analyzes the discrepancy between the data set corresponding to the third group of values ​​and the data set corresponding to the second group of values. For example, the analysis unit 17 compares the data set corresponding to a time when the operating efficiency was good (the third group of values) with the data set corresponding to the second group of values ​​to analyze which data are discrepant. The analysis unit 17 may perform the analysis using an abnormality diagnosis method assuming that the third group of values ​​represents a normal state and that the second group of values ​​represents an abnormal state.

[0051] In step S7 (output step), the output unit 18 outputs result data created using the data set corresponding to the third group of values ​​based on the deviation. For example, the output unit 18 may output the calculated average value and standard deviation for the deviated data as result data. The output unit 18 may transmit the result data to the terminal 40 or another device. The output unit 18 may display the result data on a display device provided in the driving optimization device 10 or another device. In one example, the output unit 18 may output result data such as that shown in FIG. 5.

[0052] [Hardware Configuration] Fig. 7 is a diagram showing an example of a hardware configuration related to the driving optimization system 1. Fig. 7 shows a computer 100 that functions as the driving optimization device 10. The computer 100 has a processor 101, a main memory unit 102, an auxiliary memory unit 103, a communication control unit 104, an input device 105, and an output device 106. The driving optimization device 10 is made up of one or more computers 100 that are made up of this hardware and software such as programs.

[0053] When the driving optimization device 10 is configured by multiple computers 100, these computers 100 may be connected locally or via a communication network such as the Internet or an intranet. This connection logically constructs a single driving optimization device 10.

[0054] The processor 101 is a CPU (Central Processing Unit) that executes an operating system, application programs, etc. The main memory 102 is composed of a ROM (Read Only Memory) and a RAM (Random Access Memory). The auxiliary memory 103 is a storage medium composed of a hard disk, flash memory, etc. The auxiliary memory 103 generally stores larger amounts of data than the main memory 102. The communication control unit 104 is composed of a network card or a wireless communication module. At least part of the communication function with other devices in the driving optimization device 10 may be realized by the communication control unit 104. The input device 105 is composed of a keyboard, a mouse, a touch panel, a microphone for voice input, etc. The output device 106 is composed of a display, a printer, etc.

[0055] The auxiliary storage unit 103 stores in advance a program 110 (driving optimization program) and data necessary for processing. The program 110 causes the computer 100 to execute each functional element of the driving optimization device 10. The program 110 causes, for example, processing related to the driving optimization method described above to be executed in the computer 100. For example, the program 110 is read by the processor 101 or the main storage unit 102, and causes at least one of the processor 101, the main storage unit 102, the auxiliary storage unit 103, the communication control unit 104, the input device 105, and the output device 106 to operate. For example, the program 110 reads and writes data from and to the main storage unit 102 and the auxiliary storage unit 103.

[0056] The program 110 may be provided in the form of a computer-readable storage medium, such as, but not limited to, a CD-ROM, a DVD-ROM, or a semiconductor memory. The program 110 may also be provided as a data signal via a communication network.

[0057] As described above, a driving optimization method according to one aspect of the present disclosure is executed by a computer. The driving optimization method includes: an acquisition step of acquiring a dataset indicating the operating performance of a target system 20, the dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters; an index setting step of setting a set of numerical values ​​of indicators to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the set of numerical values ​​of the indicators, a first set of numerical values ​​included in a first period that is a predetermined period and a second set of numerical values ​​included in a second period that is a portion of the first period; a target determination step of determining, using a variance in the first set of numerical values, a target value that is a variable for optimizing the second set of numerical values; a second extraction step of extracting, from the first set of numerical values, a third set of numerical values ​​that is a set of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third set of numerical values ​​and a dataset corresponding to the second set of numerical values; and an output step of outputting result data created using the dataset corresponding to the third set of numerical values ​​based on the deviation.

[0058] An operation optimization program according to one aspect of the present disclosure causes a computer to execute the following steps: an acquisition step of acquiring a dataset indicating the operating performance of a target system 20, the dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters; an index setting step of setting a group of numerical values ​​of an index to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the group of numerical values ​​of the index, a first group of numerical values ​​included in a first period, which is a predetermined period, and a second group of numerical values ​​included in a second period, which is a portion of the first period; a target determination step of determining, using the variance of the first group of numerical values, a target value that is a variable for optimizing the second group of numerical values; a second extraction step of extracting, from the first group of numerical values, a third group of numerical values ​​that is a group of numerical values ​​that achieves the target value; an analysis step of analyzing the deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step of outputting result data created using a dataset corresponding to the third group of numerical values ​​based on the deviation.

[0059] A storage medium according to one aspect of the present disclosure is a computer-readable storage medium. The storage medium has recorded thereon an operation optimization program including: an acquisition step of acquiring a dataset indicating operational performance of a target system (20), the dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters; an index setting step of setting a set of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the set of numerical values ​​of the indexes, a first set of numerical values ​​included in a first period that is a predetermined time period and a second set of numerical values ​​included in a second period that is a portion of the first period; a target determination step of determining, using a variance in the first set of numerical values, a target value that is a variable for optimizing the second set of numerical values; a second extraction step of extracting, from the first set of numerical values, a third set of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third set of numerical values ​​and a dataset corresponding to the second set of numerical values; and an output step of outputting result data created using the dataset corresponding to the third set of numerical values ​​based on the deviation.

[0060] In a driving optimization method, a driving optimization program, and a storage medium according to one aspect of the present disclosure, a set of values ​​of an index to be optimized is set using at least a portion of multiple factor data. A first set of values ​​for a first period and a second set of values ​​for a second period are extracted from the set of values ​​of the index. A target value, which is a variable for optimizing the second set of values, is determined based on the variance of the first set of values. A third set of values ​​that achieves the target value is then extracted from the first set of values. Based on the deviation between a data set corresponding to the third set of values ​​and a data set corresponding to the second set of values, result data created using a data set corresponding to the third set of values ​​is output. That is, a third set of values ​​of a subset that achieves the target value (has good efficiency) is extracted from the first set of values ​​of the entire set, and data that causes the deviation between the third set of values ​​and the second set of values ​​is analyzed from a data set including multiple parameters and multiple factor data. The result data based on the deviation can also be considered data aiming to be champion data when efficiency is good. In other words, the result data is data that can optimize the operation of the target system 20. According to the process of the present disclosure, it is not necessary to create a model such as a physical model, and therefore it is possible to reduce the calculation cost. As a result, it is possible to reduce the calculation cost of data for optimizing the operation of the target system 20.

[0061] In the first extraction step, a first group of values ​​is extracted for the entire first period, and a second group of values ​​is extracted for the period including the latest time information for the second period. In this case, a sufficient number of data points for the first group of values ​​in the entire set can be ensured, and a second group of values ​​based on the latest operation of the target system 20 can be extracted. This makes it possible to output data that optimizes the latest operation of the target system 20.

[0062] The first extraction step narrows down at least some of the parameters and factor data included in the first time period using extraction conditions based on at least some of the parameters and factor data included in the second time period, thereby extracting a first group of numerical values. In this case, the first group of numerical values ​​is narrowed down based on at least some of the parameters and factor data included in the second time period. In other words, the first group of numerical values ​​is extracted when the data set is at least partially similar to the data set of the second time period. By using such data, the accuracy of optimizing the operation of the target system 20 can be improved.

[0063] The first extraction step narrows down at least some of the parameters and factor data included in the first period, using the extraction condition that the parameters and factor data included in the first period are normal, and extracts a first group of numerical values. In this case, the first group of numerical values ​​is narrowed down based on at least some of the normal parameters and factor data. That is, the first group of numerical values ​​is extracted based on normal data. By using normal data, the accuracy of optimizing the operation of the target system 20 can be improved.

[0064] [Modifications] The present disclosure is not necessarily limited to the above-described embodiments, and various modifications are possible without departing from the gist of the present disclosure.

[0065] In the above embodiment, an example has been described in which the operation optimization method is applied to the optimization (improvement) of heat loss (or fuel consumption) of a coal-fired boiler plant, but the present invention is not limited to this. ２ The present invention may be applied to the optimization of emissions, minimization of environmentally regulated substances, minimization of power consumption of each production facility or line in a factory, improvement of yield in batch production (for example, castings, vacuum furnaces, or heat-processed products), or improvement of energy consumption of general-purpose compressors in a factory.

[0066] [Notes] The gist of the present disclosure is as follows: [1] A method for optimizing operation executed by a computer, comprising: an acquisition step of acquiring a dataset indicating the operating performance of a target system, the dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters; an index setting step of setting a group of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the group of numerical values ​​of the indexes, a first group of numerical values ​​included in a first period that is a predetermined period and a second group of numerical values ​​included in a second period that is a portion of the first period; a target determination step of determining, using a variance in the first group of numerical values, a target value that is a variable for optimizing the second group of numerical values; a second extraction step of extracting, from the first group of numerical values, a third group of numerical values ​​that is a group of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step of outputting result data created using the dataset corresponding to the third group of numerical values ​​based on the deviation. [2] The driving optimization method described in [1], wherein the first extraction step extracts the first group of numerical values ​​for the entire first period and the second group of numerical values ​​for the period including the latest time information. [3] The driving optimization method described in [1] or [2], wherein the first extraction step narrows down at least some of the parameters and factor data included in the first period using extraction conditions based on at least some of the parameters and factor data included in the second period, and extracts the first group of numerical values. [4] The driving optimization method described in any of [1] to [3], wherein the first extraction step narrows down at least some of the parameters and factor data included in the first period, and extracts the first group of numerical values, using extraction conditions that the parameters and factor data included in the first period are normal.[5] An operation optimization program that causes a computer to execute the following steps: an acquisition step of acquiring a dataset that includes a plurality of parameters and a plurality of factor data other than the plurality of parameters and indicates the operating performance of a target system; an index setting step of setting a group of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting a first group of numerical values ​​included in a first period that is a predetermined period and a second group of numerical values ​​included in a second period that is a portion of the first period from the group of numerical values; a target determination step of determining a target value that is a variable that optimizes the second group of numerical values ​​using a variance in the first group of numerical values; a second extraction step of extracting a third group of numerical values ​​from the first group of numerical values, which is a group of numerical values ​​that achieves the target value; an analysis step of analyzing the deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step of outputting result data created using the dataset corresponding to the third group of numerical values ​​based on the deviation. [6] A computer-readable storage medium having recorded thereon an operation optimization program comprising: an acquisition step of acquiring a dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters, and indicating the operating performance of a target system; an index setting step of setting a group of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the group of numerical values ​​of the indexes, a first group of numerical values ​​included in a first period that is a predetermined period, and a second group of numerical values ​​included in a second period that is a portion of the first period; a target determination step of determining, using a variance in the first group of numerical values, a target value that is a variable for optimizing the second group of numerical values; a second extraction step of extracting, from the first group of numerical values, a third group of numerical values ​​that is a group of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step of outputting result data created using the dataset corresponding to the third group of numerical values ​​based on the deviation.

[0067] REFERENCE SIGNS LIST 1 Driving optimization system 10 Driving optimization device 20 Target system 30 External system 40 Terminal 11 Acquisition unit 12 Database 13 Index setting unit 14 First extraction unit 15 Goal determination unit 16 Second extraction unit 17 Analysis unit 18 Output unit

Claims

1. A driving optimization method executed by a computer, comprising: an acquisition step of acquiring a dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters, and indicating the driving performance of a target system; an index setting step of setting a group of numerical values ​​of indexes to be optimized using at least a portion of the plurality of factor data; a first extraction step of extracting, from the group of numerical values ​​of the indexes, a first group of numerical values ​​included in a first period, which is a predetermined period, and a second group of numerical values ​​included in a second period, which is a portion of the first period; a target determination step of determining a target value, which is a variable for optimizing the second group of numerical values, using a variance in the first group of numerical values; a second extraction step of extracting, from the first group of numerical values, a third group of numerical values ​​which is a group of numerical values ​​that achieves the target value; an analysis step of analyzing a deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step of outputting result data created using the dataset corresponding to the third group of numerical values ​​based on the deviation.

2. A driving optimization method as described in claim 1, wherein the first extraction step extracts the first group of numerical values ​​using the first period as the entire period, and extracts the second group of numerical values ​​using the second period as the period including the latest time information.

3. The driving optimization method described in claim 1, wherein the first extraction step narrows down at least a portion of the multiple parameters and multiple factor data included in the first period using extraction conditions based on at least a portion of the multiple parameters and multiple factor data included in the second period, and extracts the first group of numerical values.

4. The driving optimization method described in claim 1, wherein the first extraction step narrows down at least a portion of the plurality of parameters and the plurality of factor data included in the first period, using the plurality of parameters and the plurality of factor data included in the first period as an extraction condition that the plurality of parameters and the plurality of factor data are normal, and extracts the first group of numerical values.

5. An operation optimization program that causes a computer to execute the following steps: an acquisition step for acquiring a dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters, and indicating the operating performance of a target system; an index setting step for setting a group of numerical values ​​of an index to be optimized using at least a portion of the plurality of factor data; a first extraction step for extracting, from the group of numerical values ​​of the index, a first group of numerical values ​​included in a first period, which is a predetermined period, and a second group of numerical values ​​included in a second period, which is a portion of the first period; a target determination step for determining a target value, which is a variable for optimizing the second group of numerical values, using a variance in the first group of numerical values; a second extraction step for extracting, from the first group of numerical values, a third group of numerical values ​​which is a group of numerical values ​​that achieves the target value; an analysis step for analyzing the deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step for outputting result data created using the dataset corresponding to the third group of numerical values ​​based on the deviation.

6. A computer-readable storage medium having recorded thereon an operation optimization program comprising: an acquisition step for acquiring a dataset including a plurality of parameters and a plurality of factor data other than the plurality of parameters, and indicating the operating performance of a target system; an index setting step for setting a group of numerical values ​​of an index to be optimized using at least a portion of the plurality of factor data; a first extraction step for extracting, from the group of numerical values ​​of the index, a first group of numerical values ​​included in a first period, which is a predetermined period, and a second group of numerical values ​​included in a second period, which is a portion of the first period; a target determination step for determining, using a variance in the first group of numerical values, a target value that is a variable for optimizing the second group of numerical values; a second extraction step for extracting, from the first group of numerical values, a third group of numerical values ​​that is a group of numerical values ​​that achieves the target value; an analysis step for analyzing a deviation between a dataset corresponding to the third group of numerical values ​​and a dataset corresponding to the second group of numerical values; and an output step for outputting result data created using the dataset corresponding to the third group of numerical values ​​based on the deviation.

Images