JP6249982B2 - Cost prediction apparatus and program - Google Patents

Description

  The present invention relates to a cost prediction apparatus and a program.

  A technique for recording the operating status of a device such as a home appliance and analyzing the status or suggesting a usage method for each device is known. For example, Patent Document 1 describes that the failure rate is predicted from the operation status of each device.

JP 11-120473 A

  However, with the conventional configuration, there is a problem that processing can be performed only in units of equipment, and cannot be used for applications in which costs are predicted in units of facilities (such as homes and buildings) having a plurality of devices.

  The present invention has been made to solve such problems, and an object of the present invention is to provide a cost prediction apparatus capable of predicting costs on a facility basis.

In order to solve the above-mentioned problems, the cost prediction apparatus according to the present invention is:
A cost prediction device that predicts costs based on information about equipment,
Electricity price forecasting function to predict future electricity prices,
With a failure response cost prediction function that predicts future failure response costs,
In the cost forecasting device, the electricity price forecasting function and the failure handling cost forecasting function are executed in units of facilities having a plurality of devices,
Electric fee prediction function is based on the number of years elapsed from regular and aging dependent obtained by analyzing the actual value of the power consumption of equipment to be acquired, purchased or year production year of the equipment, equipment And a function of predicting the electricity bill for the device for each of a plurality of periods based on the predicted power consumption .

The electricity rate prediction function calculates the amount of power consumption considering the aging dependence,
Power consumption that eliminates the increase in power consumption due to the effects of aging, and
Increase in power consumption due to the effects of aging
Including power consumption prediction formula including
The increase in power consumption may be obtained based on the dependence on aging.
The failure handling cost prediction function may include a function of predicting the failure handling cost of the device for each of a plurality of periods based on the age of the device and a record relating to the failure of the device.
Electricity rate prediction function or failure response cost prediction function is a post-replacement cost that calculates the electricity rate or failure response cost for another device when at least one device is replaced with another device for at least one facility. A calculation function may be included.
The replacement cost calculation function
A function for calculating the first post-replacement cost when the device is replaced in the first period;
A function for calculating the second post-replacement cost when the device is replaced in the second period;
Including
The cost prediction device further determines the future electricity rate when the replacement is made at any of the first period and the second period based on the first replacement cost and the second replacement cost. There may be a timing function that determines whether it will be smaller, future failure cost will be smaller, or the sum of future electricity charges and future failure cost will be smaller .
The failure response cost prediction function may calculate repair costs, replacement costs, or both according to variables or constants set for each facility.

  The program according to the present invention causes a computer to function as the above-described cost prediction apparatus.

  The cost predicting apparatus according to the present invention can predict a cost required in the future for a facility having a plurality of devices in units of facilities.

It is a figure which shows the example of a structure containing the expense prediction apparatus which concerns on Embodiment 1 of this invention. It is a flowchart explaining the flow of a process of the apparatus operation log data acquisition part of FIG. It is a flowchart explaining the flow of a process of the power consumption amount secular variation prediction part of FIG. It is a flowchart explaining the flow of a process of the apparatus failure rate prediction part of FIG. It is a flowchart explaining the flow of a process of the asset cost calculation part of FIG. It is a flowchart showing the detail of step S41 of FIG. It is a flowchart showing the detail of step S42 of FIG. It is a figure showing the actual value and prediction formula of the power consumption which removed the influence of aged deterioration. It is a figure which shows an example of the preparation method of a failure rate prediction type | formula. It is a correspondence table of error contents and discomfort costs. 10 is a flowchart for explaining a processing flow of an asset cost calculation unit according to the second embodiment. It is a figure which shows the example of a structure of the expense prediction apparatus which concerns on Embodiment 3. FIG. It is a flowchart explaining the flow of a process of the apparatus utilization advice preparation part of FIG.

Embodiments of the present invention will be described below with reference to the accompanying drawings.
Embodiment 1 FIG.
In FIG. 1, the example of a structure containing the expense prediction apparatus 10 which concerns on Embodiment 1 of this invention is shown.

  One or more devices (devices 30 to 36 in the example of FIG. 1) are respectively installed in at least one facility (facility 20, 21 in the example of FIG. 1). The facility may be, for example, a home, a building such as an office building, a public facility, or an apartment house such as an apartment. The device is a device that consumes power, and may be a home appliance such as an air conditioner, a refrigerator, or a rice cooker.

  Each facility is provided with at least one management device for managing devices. In the example of FIG. 1, the management devices are an energy management system 40, a network adapter 41, and an energy management system 42.

  The energy management system 40 is communicably connected to the devices 30 to 32 and transmits / receives information to / from the devices 30 to 32. Further, the energy management system 40 is connected to the communication network N, and transmits / receives information to / from an external computer (for example, the cost prediction device 10) via the communication network N. The communication network N is, for example, the Internet, but may be a VPN, a dedicated line, or the like.

  Similarly, the network adapter 41 transmits and receives information between the device 33 and the communication network N, and the energy management system 42 transmits and receives information between the devices 34 to 36 and the communication network N.

  As described above, the cost prediction apparatus 10 is configured to be able to communicate with a management apparatus (such as the energy management system 40) installed in the facilities 20 and 21 via the communication network N.

  Here, the cost prediction device 10 may have a function of registering each management device. For example, the person in charge of the facility 20 uses the site for the user of the cost prediction device 10 to confirm that there are two types of device connection routes in the facility 20 via the energy management system 40 and the network adapter 41 in advance. It may be possible to register in the cost prediction apparatus 10.

  The cost prediction apparatus 10 includes a configuration as a computer, and includes a calculation unit that performs calculation, a storage unit that stores information, an input unit that receives a user's operation, an output unit that outputs information, and an external communication network. Communication means for receiving input / output of information.

  For example, the calculation means includes a CPU (Central Processing Unit), and the storage means includes a storage medium such as a semiconductor memory and an HDD (Hard Disk Drive). The input means includes, for example, a keyboard and a mouse. The output means includes, for example, a liquid crystal display and a speaker. The communication unit may function as an input unit or an output unit.

  The calculation means of the cost prediction apparatus 10 functions as a device operation log data acquisition unit 11, an asset cost calculation unit 12, a power consumption aging change prediction unit 13, and a device failure rate prediction unit 14 by executing a program (not shown). In addition, other functions described in the present specification are realized. That is, this program causes a computer to function as the cost prediction apparatus 10.

  The storage means of the cost prediction apparatus 10 stores a device operation log database 15 and a device failure handling history database 16.

The device operation log database 15 includes operation log data and error log data for each device.
The operation log data is data that associates information for identifying a device (for example, a main product ID) with information representing an operation state of the device at a specific time or a specific period. The operating state is, for example, the age of the device (for example, the number of years since the purchase year or the number of years since the production year), the model (model) of the device, and the power state is on. It includes whether it is present or off, power consumption, device-specific parameters, and the like. The power consumption amount is a value representing the actual power consumption value of each device, that is, the actual power consumption amount. The power consumption amount may be, for example, a power consumption amount for 5 minutes or a power consumption amount for 60 minutes. For example, in the case of an air conditioner, the device-specific parameters include a set temperature, an operation mode (cooling, heating, air blowing, dehumidification, etc.), a room temperature measurement value, an outside air temperature measurement value, and the like.

  Further, the information indicating the operating state may include information indicating the state of the component. Information representing the state of the component includes, for example, a heater temperature, a compressor pressure, and the like.

  The error log data is an example of a record relating to a device failure, and stores information for identifying the device (for example, a main product ID) and information related to an error that has occurred in the device in association with each other. Information about an error is transmitted to the cost prediction device 10 from the management device or from the devices 30 to 36 via the management device every time an error occurs. Information related to the error includes the age of the device, model, date and time when the error occurred, information identifying the part in which the error occurred (for example, part ID), information identifying the content of the error (for example, error content ID), and canceling the error Date and time, etc.

  The device failure handling history database 16 includes a device failure handling history representing a history of dealing with failures of each device. The equipment failure handling history is input by, for example, an administrator of the cost prediction apparatus 10. The device failure handling history is an example of a record related to a device failure. For example, information (for example, customer ID or customer name) for identifying a user (customer) of the device and failure of the device used by the customer. Are stored in association with the history of correspondence. The history of correspondence includes information for identifying the device (for example, main product ID), correspondence date and time (for example, date and time of repair work), failure content (for example, data corresponding to the above error content ID or text data), correspondence Components (for example, those corresponding to the above-described component IDs or text data). Further, the history of correspondence may include the date and time when a customer inquiry occurred, the contents of the inquiry, and the like.

  The device operation log data acquisition unit 11 collects operation log data and error log data of the devices 30 to 36 and stores them in the device operation log database 15.

  The power consumption aging change prediction unit 13 predicts the aging deterioration tendency of the power consumption of each device, as will be described later, using the operation log data and the error log data.

  The equipment failure rate prediction unit 14 predicts a trend of the failure occurrence probability of each equipment over time using error log data, equipment failure handling history, and the like.

  The asset cost calculation unit 12 includes the aging deterioration tendency of the power consumption of each device created by the power consumption aging prediction unit 13 and the aging trend of the failure occurrence probability of each device created by the device failure rate prediction unit 14. Is used to calculate the asset cost for each facility. Asset costs include future electricity charges and future failure handling costs. The failure handling cost includes, for example, a cost related to repair, a cost related to replacement, and the like.

  That is, the asset cost calculation unit 12 predicts a future failure fee based on the electricity rate prediction function for predicting a future electricity rate based on the actual amount of power consumption of the device and a record relating to the failure of the device. It can be said that it is provided with a failure handling cost prediction function.

  In particular, the asset cost calculation unit 12 according to the present invention can calculate the asset cost for each facility. That is, it can be said that the electricity price prediction function and the failure handling cost prediction function of the asset cost calculation unit 12 can be executed in units of facilities having a plurality of devices.

  With such a configuration, the cost prediction apparatus 10 predicts a cost that will be required in the future for at least one facility 20, 21 as described below, based on information about the devices 30 to 36.

  FIG. 2 is a flowchart for explaining the processing flow of the device operation log data acquisition unit 11. The processing of FIG. 2 is started, for example, periodically (for example, every 5 minutes or every 60 minutes).

  The device operation log data acquisition unit 11 acquires device operation log data and error log data (step S11). For example, the device operation log data acquisition unit 11 receives operation log data and error log data transmitted from the energy management system 40, the network adapter 41, or the energy management system 42.

  As a modified example, when the energy management system 40 or the like does not have a function of periodically transmitting operation log data and error log data to the cost prediction apparatus 10, the energy management system 40 from the device operation log data acquisition unit 11. The operation log data and the error log data may be periodically transmitted to the user, and the operation log data and the error log data may be received as a response.

  The device operation log data acquisition unit 11 records and stores the received operation log data and error log data in the device operation log database 15 (step S12). Thereafter, the device operation log data acquisition unit 11 ends the process of FIG.

  FIG. 3 is a flowchart for explaining the processing flow of the power consumption aging change prediction unit 13. The power consumption aging change prediction unit 13 determines the prediction formula for predicting the aging change of the power consumption for each model by executing the processing of FIG. 3. However, the power consumption aging change prediction unit 13 may not execute the processing of FIG. 3, and in that case, a fixed power consumption prediction formula may be used. The fixed power consumption prediction formula is defined for each model, for example.

  In the process of FIG. 3, first, the power consumption aging change prediction unit 13 analyzes the aging dependence of the power consumption for each model of the device based on the operation log data (step S <b> 21).

  In the following, as an example, a method for analyzing the dependence of power consumption of an air conditioning facility on aged deterioration will be described. The power consumption aging change prediction unit 13 uses at least one condition pattern for the device of a specific model among the operation log data included in the device operation log database 15 and determines the condition pattern corresponding to each condition pattern. To extract everything.

  A condition pattern can be defined for each device type. For example, in the case of an air conditioner, it can be defined that the operation mode is a specific mode, the outside air temperature is included in a specific temperature range, and the set temperature is included in a specific temperature range. Specifically, a condition pattern that “the operation mode is cooling, the outside air temperature is 30 ° C. to 32 ° C., and the set temperature is 26 ° C. to 28 ° C.” can be used.

  Note that all operation log data may be extracted without using the condition pattern. In this case, all the operation log data can be regarded as corresponding to one condition pattern. The condition pattern may be defined for each month. In this way, in the power consumption prediction (step S41 in FIG. 5) to be described later, it is possible to perform processing in consideration of the season, such as obtaining the predicted value for August of the following year based on the actual value of August of a certain year. It becomes.

  Next, the power consumption aging change prediction unit 13 uses, for each condition pattern, the power consumption (actual value) in the extracted operation log data and the power consumption prediction formula of the model corresponding to the extracted operation log data. And evaluate the error.

  The power consumption aging change prediction unit 13 can acquire the power consumption prediction formula used for this evaluation by various methods. For example, when past power consumption data exists for each condition pattern, an approximate expression may be created for each condition pattern for the past power consumption data. When past power consumption data does not exist, the approximate expression for each condition pattern of the conventional model may be applied as it is, or even if the power consumption change tendency for each condition pattern of general air conditioning equipment is applied. Good. At this time, the same prediction formula may be acquired for a plurality of condition patterns (or for all condition patterns).

As specific examples of the power consumption prediction formula, the following formulas 1 to 3 can be considered. In any case, E _f is a power consumption amount, E _p is a constant that can be different for each condition pattern (for example, E _p = 1000 for a specific condition pattern), and W is an elapsed year. In addition, a and b in Equation 3 are constants that can be different for each condition pattern.
-Linear approximation, for example, E _f = E _p + 10 × W [kWh] (Equation 1)
Exponential approximation, for example E _f = E _p × e ^{0.003 × W} [kWh] (Equation 2)
-Polynomial approximation, for example, E _f = a × W ² + b × W + E _p [kWh] (Equation 3)

  As a modification, one or more of variables defining the condition pattern may be used as a variable of the power consumption prediction formula, and may be defined as the following formula 4, for example.

... (Formula 4)
However _{T 1} is the outside temperature [℃], _{T 2} is the set temperature [° C.]. If it does in this way, one prediction formula can be shared considering a specific condition about a plurality of condition patterns.

  The power consumption aging change prediction unit 13 then obtains the theory obtained from the power consumption prediction formula based on the actual value of the power consumption, the model related to the operation log data, and the elapsed years for each of the extracted operation log data. Based on the value, an error between the actual value and the theoretical value is calculated for each condition pattern. The error may be obtained by subtraction, for example, or may be obtained by another calculation.

  In this way, the power consumption aging change prediction unit 13 determines an error for each condition pattern for all the operation log data for all the condition patterns. Then, the power consumption aging change prediction unit 13 determines whether or not this error is equal to or greater than a predetermined reference (step S22). The comparison between the error and the reference can be performed, for example, by comparing the standard deviation of the error with a predetermined threshold, but other known methods may be used.

  If the error with respect to a certain condition pattern is less than the reference, the power consumption aging prediction unit 13 ends the process of FIG. 3 for that condition pattern. On the other hand, if the error with respect to a certain condition pattern is equal to or greater than the reference (for example, the standard deviation is equal to or greater than the threshold), the power consumption aging change prediction unit 13 corrects the power consumption prediction formula of the condition pattern of the model (step S23). ).

  The prediction formula can be corrected by various methods. For example, when the prediction formula is an approximate formula created based on past power consumption data, by changing the population of data that is the basis of the approximate formula (for example, excluding relatively old data) The parameter identification of the prediction formula may be performed again. In addition, a new prediction formula may be created using a function model defined in advance, such as linear approximation, least squares, or regression analysis.

  When the correction of the prediction formula is completed, the power consumption aging prediction unit 13 updates the power consumption prediction formula related to each model and each condition pattern to the corrected prediction formula (step S24). For the update operation, a prediction formula database may be held and the prediction formula stored therein. Then, the power consumption aging change prediction unit 13 ends the process of FIG. Thus, the prediction formula for predicting the secular change of the power consumption is determined for each model.

  FIG. 4 is a flowchart for explaining the processing flow of the device failure rate prediction unit 14. The device failure rate prediction unit 14 determines the prediction formula for predicting the secular change of the failure rate for each model of the device by executing the processing of FIG. However, the equipment failure rate prediction unit 14 may not execute the processing of FIG. 4, and in that case, a fixed failure rate prediction formula may be used. The fixed failure rate prediction formula is defined for each model, for example.

  In the process of FIG. 4, first, the device failure rate prediction unit 14 analyzes the dependence of the failure rate over time for each model of the device based on the error log data or the device failure handling history (step S31).

  Hereinafter, as an example, a method for analyzing the dependence of the failure rate of the air conditioning equipment on the aging will be described. The device failure rate prediction unit 14 extracts all of the error log data included in the device operation log database 15 related to a specific model device. Then, the extracted error log data is classified for each elapsed year, and a failure rate is calculated for each elapsed year.

  The failure rate is calculated by, for example, calculating the number of detected failures (for example, the number of records in the extracted error log data) and the number of devices of the corresponding model (for example, the device operation log database 15 including the operation log data). Divided by the total number of devices). Alternatively, the number of detected failures may be calculated by dividing by the number of units shipped to the market.

  Next, the equipment failure rate prediction unit 14 evaluates an error between the calculated failure rate (actual value) for each elapsed year and the failure rate prediction formula of the model corresponding to the extracted operation log data.

  The equipment failure rate prediction unit 14 can acquire the failure rate prediction formula used for this evaluation by various methods. For example, when past failure rate data exists, an approximate expression may be created for the past failure rate data. If past failure rate data does not exist, the approximation formula of the conventional model may be applied as it is, and the failure rate prediction formula of general air conditioning equipment known from literature etc. is common to multiple models You may apply to.

  The equipment failure rate prediction unit 14 then, for each of the extracted operation log data, based on the actual value of the failure rate and the theoretical value obtained from the failure rate prediction formula based on the model and the elapsed years related to the error log data. The error between the actual value and the theoretical value is calculated. The error may be obtained by subtraction, for example, or may be obtained by another calculation.

  In this way, the equipment failure rate prediction unit 14 determines an error for each elapsed year. Then, the equipment failure rate prediction unit 14 determines whether this error is equal to or greater than a predetermined reference (step S32). The comparison between the error and the reference can be performed, for example, by comparing the standard deviation of the error with a predetermined threshold, but other known methods may be used.

  If the error is less than the reference, the equipment failure rate prediction unit 14 ends the process of FIG. If the error is greater than or equal to the reference (for example, the standard deviation is greater than or equal to the threshold), the equipment failure rate prediction unit 14 corrects the failure rate prediction formula of the model (step S33).

  The prediction formula can be corrected by various methods. For example, when the prediction formula is an approximate formula created based on past failure rate data, by changing the data set that is the basis of the approximate formula (for example, excluding relatively old data), The parameter identification of the prediction formula may be performed again. In addition, a new prediction formula may be created using a function model defined in advance, such as linear approximation, least squares, or regression analysis.

  When the correction of the prediction formula is completed, the device failure rate prediction unit 14 updates the failure rate prediction formula for each model to the corrected prediction formula (step S34). For the update operation, a prediction formula database may be held and the prediction formula stored therein. Then, the equipment failure rate prediction unit 14 ends the process of FIG.

  5 to 7 are flowcharts for explaining the flow of processing of the asset cost calculation unit 12. In particular, FIG. 6 shows details of step S41 in FIG. 5, and FIG. 7 shows details of step S42 in FIG.

  The asset cost calculation unit 12 predicts future device-related expenditures for each facility by executing the processes of FIGS. Here, the equipment-related expenditure means expenditure generated in relation to the equipment installed in the facility, and includes future electricity charges and future failure handling expenses.

  In the process of FIG. 5, first, the asset cost calculation unit 12 predicts a future electricity rate for each facility (step S <b> 41). Details of step S41 will be described later with reference to FIG. Subsequently, the asset cost calculation unit 12 predicts a future failure handling cost for each facility (step S42). Details of step S42 will be described later with reference to FIG.

  Subsequently, the asset cost calculation unit 12 predicts future equipment-related expenditure in units of facilities by adding them based on the predicted electricity charge and the predicted failure handling cost (step S43). ).

  Next, the asset cost calculation unit 12 outputs the predicted device-related expenditure (step S44). This output is performed via the communication network N, for example. For example, the equipment-related expenditure predicted for the facility 20 may be transmitted to the energy management system 40 or may be transmitted to the network adapter 41.

  In addition, the equipment-related expenditure may be output according to the HTTP protocol. For example, device-related expenditure may be stored in association with a specific URL, and may be transmitted in response to the person in charge of the facility 20 accessing this URL. In this case, the equipment-related expenditure is output to a screen of a web browser installed in a mobile communication terminal or a PC owned by the person in charge of the facility 20, for example. Then, the asset cost calculation part 12 complete | finishes the process of FIGS.

  FIG. 6 shows details of step S41 in FIG. In the process of FIG. 6, first, the asset cost calculation unit 12 acquires a power consumption amount prediction formula defined by the process of FIG. 3 or other processes for each model to be processed (step S411). Subsequently, for each device to be processed, the past power consumption (actual value) is acquired from the operation log data of the device operation log database 15 (step S412).

  It should be noted that a method for determining which model and which device is the target of step S41 can be appropriately designed by those skilled in the art. For example, the target model or device may be stored in advance in the storage unit of the cost prediction apparatus 10. Alternatively, all devices included in the operation log data and all models of these devices may be targeted.

Then, asset cost calculation unit 12 is calculated by using the power consumption of the past which has obtained the (actual value), per unit time, power consumption E _p removing the effects of aging (step S413). The length of the unit period may be arbitrarily set by an administrator or user of the cost prediction apparatus 10.

  Hereinafter, in the present embodiment, the unit period is described as one month. In this case, the asset cost calculation part 12 calculates | requires the cumulative value of the power consumption (actual value) for one month about each apparatus. As a modification, the power consumption amount may not be obtained as a cumulative value. For example, the power consumption amount may be obtained as a daily load curve when a charge by time zone is applied as the contract form of the electricity charge or when it is desired to accurately calculate the contract power.

The amount of power consumption from which the influence of aging degradation has been removed can be calculated as follows, for example.
First, the asset cost calculation unit 12 extracts all the air conditioners that correspond to a certain condition pattern from the operation log data. Then, assuming that the operation continues for one month (for example, every day for a predetermined time) with the condition pattern, a cumulative value of power consumption for one month is calculated. As a specific example of the cumulative value calculation method, it is assumed that there are 10 hours of operation log data corresponding to the condition pattern, the total power consumption is 5 kWh, and one month is 30 days × 8 hours. Then, the cumulative value of power consumption for one month is 5 ÷ 10 × 30 × 8 = 120 [kWh].

In addition, the asset cost calculation unit 12 uses the power consumption prediction formula of each model and each condition pattern, and removes the power consumption increase E due to the influence of aged deterioration corresponding to the elapsed years of each device. _p is calculated. The E _p is a value representing the power consumption of the time elapsed years is zero. This calculation is not necessary when _Ep is a constant in the prediction formula. When E _p is a variable in the prediction formula, for example, E _p can be determined as follows.

For example, suppose that the power consumption prediction formula is E _f = E _p + 5 × W [kWh] for a certain air-conditioning equipment model. In addition, the purchase year of the air conditioning equipment with this model is 2010, and the actual value (cumulative value for one month) E _f of the power consumption when the elapsed time W is 2 [years] is 1015 [kWh ]. This numerical value is considered to be a numerical value as a result of an increase in power consumption due to the influence of aging for two years.

Substituting actual values of the air conditioning equipment to the prediction _{equation, 1015 = E p + 5 ×} 2 becomes, _E p = 1005 is obtained Solving for _{E p.}

And the asset cost calculation part 12 estimates the power consumption per unit period which considered the influence of aged deterioration for every elapsed years about each year including a future period (step S414). For example, when the prediction formula is obtained as E _f = 1005 + 5 × W [kWh] as described above, the future E _f is calculated for each of the elapsed years W = 0, 1,. However, the calculation is omitted for W in which all of the corresponding year belongs to the past. N represents the maximum number of years that should be predicted.

Here, a specific example of the flow of processing from the creation of the prediction formula to the future prediction is summarized as follows.
FIG. 8 is a diagram illustrating the actual value of power consumption and the prediction formula from which the influence of aging deterioration has been removed. As shown by the “□” symbol in FIG. 8, when there is power consumption data from which the influence of the aging deterioration of the same month (that is, the same condition pattern) for the past five years is removed, the influence of the past aging deterioration is removed. A correlation formula (prediction formula) for the power consumption can be created, and a value indicated by a “◯” symbol can be obtained as the power consumption from which the influence of future aging is removed. The correlation equation (prediction equation) can be created by using a general method such as linear approximation or least square method. Subsequently, the power consumption amount from which the influence of the future aging predicted in step S413 is removed is substituted into the power consumption prediction formula considering the influence of the aging deterioration as shown in the above formula 1, and the influence of the aging deterioration is obtained. The future power consumption per unit period considering the above is calculated (that is, the future power consumption appears above the solid line representing the prediction formula in the graph of FIG. 8). In this manner, the process of calculating the power consumption considering the influence of aging deterioration is repeated for each model unit, for each condition pattern, and for each elapsed year.

  When the calculation of the power consumption considering the influence of future aging is completed, the asset cost calculation unit 12 considers all the devices installed in the facility for each condition pattern and for each elapsed year for each facility. The future electricity charge per unit period is predicted (step S415, electricity charge prediction function). The prediction of the electricity bill is performed based on the power consumption calculated in step S414.

  The electricity price may be predicted in any way, but an example will be described below. In this example, the calculation method of the electricity charge differs depending on the power contract form for each facility. In the contract form of a certain facility, a fixed electricity charge is applied at all times. In such a facility, the electricity charge per unit period is the sum of all future power consumption per unit period of the equipment placed in the facility, and this sum is multiplied by a predetermined ratio (unit price of electricity charge). Is calculated by

  Further, in another facility contract form, an inexpensive power charge unit price is applied up to a predetermined power consumption upper limit value, and an expensive power charge unit price is applied to a portion exceeding the upper limit value. In such a facility, first, all the future power consumption per unit period of the equipment placed in the facility is totaled, and the portion below the upper limit is multiplied by the cheap unit price of the electricity bill, For the portion exceeding, the electricity bill is calculated by multiplying the expensive electricity bill unit price and calculating the sum of these.

  If there are multiple condition patterns, the time of operation log data corresponding to each condition pattern is integrated for each condition pattern, and the power consumption predicted for each condition pattern is calculated according to the integrated value of time. You may weight and total.

  Next, the asset cost calculation unit 12 outputs the calculated future electricity bill (step S416). The output destination may be the storage means of the cost prediction apparatus 10 or may be the communication network N or another communication network (in this case, a computer or the like that has received this via the communication network N or the like). , You may have an electricity price prediction result database and store it there). In this way, the asset cost calculation unit 12 ends the process of FIG. 6, that is, step S41 of FIG.

  Here, when the power consumption prediction formula is created based on the actual value of power consumption, the electricity rate prediction function in step S415 performs the elapsed years of the device and the actual power consumption of the device. Based on the value, it can be said that it includes a function of predicting the electricity bill of the device for each of a plurality of periods.

  FIG. 7 shows details of step S42 of FIG. In the process of FIG. 7, the asset cost calculation unit 12 first acquires a failure rate prediction formula defined by the process of FIG. 4 or other processes for each model to be processed (step S421). Subsequently, for each device to be processed, the past error log data is acquired from the device operation log database 15 and the device failure response history is acquired from the device failure response history database 16 (step S422).

  The asset cost calculation unit 12 may use the failure rate prediction formula for each model in common for all devices of the model, or create different failure rate prediction formulas for each device based on the failure rate prediction formula of the model. However, this may be used for each device.

  FIG. 9 shows an example of a method for creating a failure rate prediction formula that differs for each device. The asset cost calculation unit 12 creates prediction formulas F1 and F2 for each device by applying different corrections for each device based on the prediction formula F for each model. For example, if the number of error log data for a device or the frequency of occurrence per unit period is less than the reference value for that model (for example, the average value of all devices of that model), the failure rate for that device will be low Thus, the prediction formula F is corrected to be the prediction formula F1.

  Further, for example, when the error log data satisfies a specific condition (for example, there are a predetermined number or more of places where the occurrence location has a high probability of failure or repair), the failure rate is determined accordingly. The prediction formula F is corrected so as to change to the prediction formula F2 (for example, it is corrected so that the failure rate becomes high). Such a condition can be determined based on the component ID and error content ID included in the error log data.

  For example, when the equipment is an air conditioner, a part ID of a predetermined ratio or more in the error log data does not directly affect the operation of the main body such as a remote controller, or an error content ID of a predetermined ratio or more is easily contaminated by the user. If the content is compatible with the above, the failure rate prediction formula may not be corrected. On the other hand, if part IDs of a certain percentage or more in the error log data are generated at locations that can only be repaired by specialists such as outdoor unit fans and refrigerant gas, the failure rate prediction formula can be used to increase the failure rate. Correction may be performed.

  The following can be used as a specific example of the prediction formula.

Where t is the number of years elapsed, y (t) is the failure rate, and c ₀ , c ₁ , t ₀ and λ are constants.

  Subsequently, the asset cost calculation unit 12 calculates a replacement cost when the device is replaced (step S423). The replacement cost is an aspect of the failure handling cost, and is a cost required when a certain device is replaced with another device. The replacement cost may be calculated by any method. For example, the cost prediction apparatus 10 may store the sales price of each model in advance, and this sales price may be used as a replacement cost.

Further, the replacement cost may be calculated according to the following equation 5, for example.
C = αC _p + βL _s + γL _UC + δ (Formula 5)
However, C is a replacement cost. C _p is the purchase cost of the replacement product, L _s is the loss due to the failure of the device before replacement, and L _UC is the amount of unpleasant cost from the failure of the device to the purchase of the replacement product These are converted values, and α, β, γ, and δ are cost increase factors associated with difficulty in obtaining substitutes.

The purchase cost C _p of the substitute product is calculated from the market price of the corresponding model or equivalent model currently in use. C _p may include a product price, installation work cost, and the like. C _p may be obtained by multiplying the number of years of purchase or the price drop coefficient. Further, C _p may be input in advance as a specification of a product that the user wants to purchase next.

Examples of the loss L _s associated with the failure of the device before replacement include expenses required due to the failure of the device. For example, in the case of a refrigerator failure, food loss inside the refrigerator, an increase in restaurant expenses, etc. can be considered. In addition, in the customer collection facilities, etc., it is conceivable that the number of customers will decrease due to the failure of the air conditioning equipment. Further, L _s may be a value obtained by multiplying the part purchase amount or the repair cost by the failure rate.

A value L _{UC obtained} by converting an unpleasant cost from the occurrence of a device failure to the purchase of a substitute product into an amount of money may be calculated by a correspondence table with error contents for each product, for example, as shown in FIG. As shown in the following formula 6, the product unit price may be multiplied by the probability of “no problem with the product” acquired through a questionnaire or the like. Further, other cost quantification methods may be used.
L _UC = C _p × P (Formula 6)
However, P is a probability that there is no problem without the product, and 0 ≦ P ≦ 1. P can be designated in advance by the person in charge of each facility, for example.

  The cost increase coefficients α, β, γ, and δ associated with the difficulty of obtaining substitute products may be calculated, for example, by using a correspondence table with general shortage times and sale times for each model. Further, the cost increase coefficient for the most recent year may be calculated based on production troubles such as natural disasters, inventory status of manufacturers, and the like.

The correspondence table in FIG. 10 associates the parts constituting the device with the unpleasant cost L _UC caused by the failure of the part. Alternatively, it can be said that the correspondence table in FIG. 10 associates the failure content of the device with the unpleasant cost L _UC generated by the failure content. This correspondence table is provided for each facility, for example. In this case, it can be said that the failure-corresponding cost prediction function of the asset cost calculation unit 12 can be executed using a different standard for each facility for prediction of replacement cost C. The correspondence table may be provided for each model, or may be provided for each device when a plurality of devices of the same model exist in the facility.

  This correspondence table may be stored in the storage means of the cost prediction device 10, and is stored in any of the management devices (energy management system 40, network adapter 41, energy management system 42, etc.) of each facility and the cost prediction device 10 May be sent to.

By using such a correspondence table, the person in charge of the facility can arbitrarily set the discomfort cost _LUC of each part or each failure content according to the state of the facility. In this case, replacement cost C for each device is calculated based on the setting for each facility. For example, even if it is the same part of the same model, when the trouble level in the case of failure differs depending on the facility, such a situation is taken into consideration, and a replacement cost C that is different for each facility is calculated. Therefore, the replacement cost C of each device can be adjusted in accordance with the situation of the facility, and as a result, the intention of the person in charge of the facility can be reflected in the prediction of the replacement cost C (and the failure handling cost).

Further, as described above, even when the uncomfortable cost L _UC is determined without using the correspondence table (for example, when L _UC = C _p × P as in Expression 6), the calculation formula is similarly applied. By setting this variable or constant for each facility, the unpleasant cost L _UC can be adjusted for each facility. Further, as long as it is a variable or a constant that defines the replacement cost C, things other than the unpleasant cost _LUC may be made different for each facility. In this case, the replacement cost C may be determined according to the above-described formula 5, or may be determined based on another formula or method.

  Subsequently, the asset cost calculation unit 12 predicts the failure occurrence time of each device (step S424). The failure occurrence timing of each device may be predicted in any way, but for example, it may be predicted in units of a period having the same length as the unit period of power consumption prediction. Further, for example, the failure rate at each time may be multiplied by a random number R (0 ≦ R ≦ 1), and when the result of multiplication exceeds a predetermined threshold, it may be predicted that a failure will occur at that time. This random number R is a random number that can take a different value for each execution of step S424 and for each period. Further, for example, based on the failure rate prediction formula, the failure occurrence rates after the current time may be integrated, and it may be determined that a failure occurs when the integration result exceeds a predetermined threshold.

  Subsequently, the asset cost calculation unit 12 predicts the repair cost of each device (step S425). The repair cost is an aspect of the failure handling cost. For example, the asset cost calculation unit 12 randomly determines the failure content. The failure content may be determined on the assumption that all the components fail with equal probability, or the failure rate of each component may be determined based on the error log data. In the case where the repair cost differs for each failure content, the repair cost may be predicted in any way. For example, a constant different for each failed part may be stored in the cost prediction apparatus 10 in advance.

  Moreover, you may enable it to set the variable or constant used for prediction of repair expense for every facility. In this case, it can be said that the failure-corresponding cost prediction function of the asset cost calculation unit 12 can be executed using a different standard for each repair facility. The variable or constant used for predicting the repair cost may be provided for each model, or may be provided for each device when a plurality of devices of the same model exist in the facility.

  Such variables or constants may be stored in the storage means of the cost prediction device 10, or stored in any facility management device (energy management system 40, network adapter 41, energy management system 42, etc.) and cost. It may be transmitted to the prediction device 10.

  By doing so, the person in charge of the facility can arbitrarily set the repair cost of each device according to the situation of the facility. In that case, the repair cost of each device is calculated based on the setting for each facility. For example, even if the same model is the same part, if the repair cost varies depending on the facility due to the presence or absence of security, such a situation is taken into consideration, and a different repair cost is calculated for each facility. Therefore, the repair cost of each device can be adjusted in accordance with the situation of the facility, and as a result, the intention of the person in charge of the facility and the like can be reflected in the prediction of the repair cost (and failure handling cost).

  Next, the asset cost calculation unit 12 predicts a failure handling cost for each facility (step S426, failure handling cost prediction function). The prediction of the failure handling cost is performed based on the repair cost and the replacement cost, for example, as follows.

  The asset cost calculation unit 12 determines, for each time when a failure is predicted to occur in step S424, whether the failure is dealt with by repair or replacement. This determination may be made at random based on, for example, a fixed probability. Or you may carry out based on the electricity bill estimated about the time. For example, it may be determined that replacement is made when the predicted electricity rate exceeds a predetermined threshold, and that repair is made otherwise.

  If it is determined that repairs are to be made, repair costs are added as failure handling costs at that time. Similarly, when it is determined that replacement is performed, replacement cost is added as a failure handling cost at that time.

  In this way, the asset cost calculation unit 12 calculates a failure handling cost for each period for each device. Thereafter, the asset cost calculation unit 12 calculates the failure handling cost for each facility by adding up the failure handling costs for the devices arranged in the facility for each period. The failure handling cost is calculated over a predetermined period (for example, 10 years from the present time).

  Subsequently, the asset cost calculation unit 12 outputs a future failure handling cost for each facility (step S427). The output destination may be the storage means of the cost prediction apparatus 10 or may be the communication network N or another communication network (in this case, a computer or the like that has received this via the communication network N or the like). , You may have a failure cost estimate database and save it there). In this way, the asset cost calculation unit 12 ends the process of FIG. 7, that is, step S42 of FIG.

  Here, when the failure rate prediction formula is created based on the error log data and other records relating to the failure, the failure-corresponding cost prediction function in step S426 performs the elapsed years of the device and the record relating to the failure of the device. Based on the above, it can be said that it includes a function of predicting the failure handling cost of the device for each of a plurality of periods.

As described above, the cost prediction apparatus 10 according to the first embodiment is a cost prediction apparatus that predicts a cost based on information related to a device, and predicts a future electricity charge. In the cost prediction apparatus, the electricity charge prediction function and the failure support expense prediction function are executed for each facility having a plurality of devices, and the electricity charge prediction function is provided. includes aging dependence obtained by analyzing the actual value of the power consumption of equipment to be acquired periodically, based on the number of years elapsed from the purchase or year production year of the equipment, the power consumption of the device predict, based on the power consumption amount predicted, the electricity charge of the equipment, because it includes the ability to predict for each of a plurality of periods, the facility having a plurality of devices, the future with respect to equipment The equipment-related expenditure which is the cornerstone, can be predicted in the accommodation units. A result of this, for example, each facility, it is possible to predict the future of equipment-related expenditure, to reduce the number of times that forced the expense of the unexpected.

In the first embodiment, the following modifications can be made.
If it is determined in step S426 that the failed device is to be replaced, the electricity rate or the failure handling cost for the unit period after the replacement time (or after that time) is predicted based on the replaced device. It may be changed. That is, for each of the devices determined to be replaced, step S41 (FIG. 6) or step S42 (FIG. 7) may be re-executed after the replacement period based on the model after replacement.

  The model after replacement may be the same as that before replacement, the current top runner model, or a model determined based on known data such as a technology trend road map. The top runner model may be defined in advance in the storage means of the cost prediction device 10 for each device managed by the cost prediction device 10 (or for each model).

  In such a modified example, the electricity rate prediction function in step S41 or the failure-corresponding cost prediction function in step S42 is performed when the at least one facility is replaced with another device. It can be said that a function (calculation function after replacement) for calculating an electricity bill or a failure handling cost is included.

  In this way, the accuracy of calculation of equipment-related expenditures due to replacement by purchase is improved, and future cost prediction regarding the equipment owned by each facility becomes more accurate.

Embodiment 2. FIG.
In the second embodiment, the operation of the asset cost calculation unit 12 is partially changed in the first embodiment, and information for determining the replacement time of the device is output.
FIG. 11 is a flowchart for explaining the processing flow of the asset cost calculation unit according to the second embodiment. The asset cost calculation unit predicts a future device-related expenditure for each facility by executing the processing of FIG.

  In the process of FIG. 11, first, the asset cost calculation unit predicts a future electricity charge, a future failure handling cost, and a future equipment-related expenditure for each facility (steps S51 to S53). These processes can be executed, for example, in the same manner as in the first embodiment (steps S41 to S43 in FIG. 5).

  Subsequently, the asset cost calculation unit determines whether the equipment-related expenditure of the facility is equal to or greater than a predetermined standard (step S54). As the predetermined standard, it may be used whether the equipment-related expenditure for a certain period exceeds a preset upper limit value. Or you may use whether the period when apparatus related expenditure exceeds the preset threshold value continuously generate | occur | produces over the predetermined period number or more.

  If the device-related expenditure is equal to or higher than the reference, the asset cost calculation unit adjusts the replacement time for the device for one or more devices in advance (step S55). “Adjust to move forward” means, for example, that if it is determined that replacement of the device is to be performed in the prediction of device-related expenditure executed immediately before (step S53), the replacement time is changed earlier. Including doing. In addition, for example, when it is determined that replacement of the device is not performed in the prediction of the device-related expenditure executed immediately before (step S53), it is determined to replace the device at any time. .

  Here, a person skilled in the art can appropriately design a device to be brought forward and a method for determining when replacement is to occur. For example, for equipment with the largest total of future electricity charges or future failure handling costs, the replacement time is advanced by one unit period (if it is decided not to replace, the replacement will be performed at the future replacement time. May be determined).

  After adjusting the replacement time in this way, the asset cost calculation unit returns the process to step S52 and recalculates the failure handling cost. (As a modification, the process may be returned to step S51 and the electricity bill may be recalculated.)

  In step 54, when the equipment-related expenditure is less than the standard (including the case where the equipment-related expenditure is less than the standard by the adjustment in step S55), the asset cost calculation unit outputs the predicted equipment-related expenditure (step S56). ). This process can be executed, for example, in the same manner as in the first embodiment (step S44 in FIG. 5). Thereafter, the asset cost calculation unit ends the process of FIG.

  Although not particularly shown in FIG. 11, the standard for ending the loop of steps S52 to S55 can be appropriately modified by those skilled in the art. For example, if the device-related expenditure does not become less than the reference even if the determination in step S54 is performed for all replacement times of all devices, the asset cost calculation unit ends the processing in FIG. 11 and is the same as in the first embodiment. 5 may be executed.

  As described above, in the second embodiment, the electricity price prediction function in step S51 or the failure response cost prediction function in step S52 is performed when at least one device is replaced with another device for at least one facility. It can be said that a function for calculating an electricity bill or a failure handling cost for another device (cost calculation function after replacement) is included.

As described above, the cost predicting apparatus according to the second embodiment is a cost predicting apparatus that predicts the cost based on the information about the device, as in the first embodiment, and predicts the future electricity bill. It has an electricity bill prediction function and a failure handling cost prediction function that predicts future failure handling costs. In the cost forecasting device, the electricity bill forecasting function and the failure handling cost forecasting function are based on facilities having a plurality of devices. The electricity price forecasting function that is executed is based on the aging dependence obtained by analyzing the actual value of the power consumption of equipment that is acquired periodically and the number of years that have elapsed since the purchase or production year of the equipment. Te, predicts the consumed electric power amount of equipment, based on the power consumption amount predicted, the electricity charge of the equipment, because it includes the ability to predict for each of a plurality of periods, the facility having a plurality of devices The equipment requirements spending future with respect to equipment, can be predicted by the accommodation units.

In the second embodiment, the following modifications can be made.
Modification 1 of Embodiment 2
The asset cost calculation unit, for all the devices placed in one facility, for all the time when each device can be replaced, the equipment-related expenditure (electricity charge or failure handling cost when the device is replaced at that time) Or the sum of both) may be predicted to determine when the power consumption is minimized. As the power consumption here, for example, the total from the unit period including the present time to the most predictable unit period that can be predicted may be used.

  In addition, the asset cost calculation unit may output information representing the device and time related to the determined combination in step S56.

  In such a configuration, the post-replacement cost calculation function includes a function for calculating the first post-replacement cost when the device is replaced at the first time, and a function when the device is replaced at the second time. A second post-replacement cost calculation function. In such a configuration, the cost prediction device replaces the first time period and the second time period based on the first post-replacement cost and the second post-replacement cost. In addition, it can be said that it has a timing function that determines whether the future electricity bill will be smaller, the future failure cost will be smaller, or the future equipment-related expenditure will be smaller. .

  If it does in this way, the effect which operates apparatus-related expenditure efficiently will be acquired, improving the energy-saving performance per facility.

Modification 2 of Embodiment 2
In the prediction of equipment failure handling costs, the cost forecasting device considers the lifestyle and life stage of the home, etc., as well as the social technology trend, scale, family structure, etc. Considering the introduction status of household devices, a model with higher performance may be proposed at the time of replacement, or additional devices may be proposed when there is room for device-related expenditure. As an example, it is assumed that the family composition is composed of four persons: a father, a mother, a child (4 years old), and a child (0 years old). Then, five to six years after the children are ready to eat, they propose to replace the refrigerator and rice cooker with a large-capacity type.

  In this way, in order to improve the quality of life / value of a facility such as a home, an effect of making a device introduction proposal accompanied by an increase in device-related expenditure can be obtained.

Embodiment 3 FIG.
In the third embodiment, the cost prediction apparatus 10 is changed to output advice on how to use the device in the first or second embodiment.

  In FIG. 12, the example of a structure of the expense prediction apparatus 110 which concerns on Embodiment 3 is shown. The cost prediction device 110 includes a device utilization advice creating unit 17 in addition to the configuration of the cost prediction device 10 (FIG. 1) of the first embodiment.

FIG. 13 is a flowchart for explaining a processing flow of the device utilization advice creating unit 17 according to the third embodiment.
In the process of FIG. 13, first, the device utilization advice creating unit 17 acquires a failure rate prediction formula for each device from a prediction formula database or the like (step S61). Subsequently, the device utilization advice creating unit 17 acquires the error log data and the failure handling history of each device from the device operation log database 15 and the device failure handling history database 16 (step S62).

  Subsequently, the device utilization advice creating unit 17 acquires operation log data of the prediction target device from the device operation log database 15 (step S63). Subsequently, the correlation between each parameter included in the operation log data, the error log, and the failure handling history is acquired (step S64).

  A method for obtaining the correlation can be appropriately designed by those skilled in the art. For example, a data analysis method such as statistical analysis or machine learning may be used. For example, for the air conditioning equipment, a correlation is obtained that there is a high correlation between the error content “compressor failure” and the condition pattern “the difference between room temperature and outside temperature is 10 ° C. or more”.

  Subsequently, the device utilization advice creating unit 17 outputs the acquired correlation (step S65). This output may be performed similarly to step S44 of FIG. Further, this correlation can be output as advice on how to use the device (device utilization advice). For example, for a facility that has an air conditioner that has an error content of “compressor failure”, “If the set temperature is changed so that the difference between the room temperature and the outside temperature is less than 10 ° C., the air conditioner May be output ”. And the apparatus utilization advice preparation part 17 complete | finishes the process of FIG.

As described above, the cost predicting apparatus according to the third embodiment is a cost predicting apparatus that predicts the cost based on the information about the device, as in the first and second embodiments, and predicts the future electricity price. An electricity rate prediction function and a failure response cost prediction function for predicting future failure handling costs. In the cost prediction device, the electricity rate prediction function and the failure response cost prediction function are provided for facilities having a plurality of devices. runs as a unit, the electric fee prediction function, and age-dependent obtained by analyzing the actual value of the power consumption of equipment to be periodically acquired, and the number of years elapsed from the purchase or year production year of the equipment based on, predicts the consumed electric power amount of equipment, based on the power consumption amount predicted, the electricity charge of the equipment, because it includes the ability to predict for each of a plurality of periods, the facility having a plurality of devices With, the equipment-related expenditure that is necessary in the future in relation to the equipment, can be predicted in the accommodation units.

  Further, according to the cost prediction apparatus 110 according to the third embodiment, it is possible to advise a long-term use parameter for a facility that desires long-term use of the possessed device, thereby improving the satisfaction with the device. An effect is obtained.

  10,110 Cost prediction device, 20,21 facilities.

Claims (7)

A cost prediction device that predicts costs based on information about equipment,
Electricity price forecasting function to predict future electricity prices,
With a failure response cost prediction function that predicts future failure response costs,
In the cost prediction apparatus, the electricity rate prediction function and the failure handling cost prediction function are executed in units of facilities having a plurality of devices ,
The electric fee prediction function is based on the number of years elapsed from regular and aging dependent obtained by analyzing the actual value of the power consumption of equipment to be acquired, purchased or year production year of the equipment, A cost prediction apparatus including a function of predicting the power consumption of the device and predicting the electricity bill of the device for each of a plurality of periods based on the predicted power consumption .   The electricity rate prediction function is configured to calculate the amount of power consumption in consideration of the aging dependency.
    Power consumption that eliminates the increase in power consumption due to the effects of aging, and
    Increase in power consumption due to the influence of the aging and
  Including power consumption prediction formula including
  The increase in power consumption is determined based on the aging dependence.
The cost prediction apparatus according to claim 1, wherein: Before Symbol failure countermeasure cost prediction functions, the number of years elapsed device, based on the record of failure of the equipment, including the ability the failure countermeasure cost of the equipment, predict for each of the plurality of periods, claim 1 Or the cost estimation apparatus of 2.   The electricity rate prediction function or failure response cost prediction function calculates the electricity rate or failure response cost for another device when at least one device is replaced with another device for at least one facility. The expense prediction apparatus as described in any one of Claims 1-3 containing the cost calculation function after replacement. The replacement cost calculation function is
A function for calculating the first post-replacement cost when the device is replaced in the first period;
A function for calculating the second post-replacement cost when the device is replaced in the second period;
Including
The cost predicting apparatus, when further replaced with any one of the first time and the second time based on the first replacement cost and the second replacement cost, Determine whether the future electricity bill is smaller, the future failure handling cost is smaller, or the sum of the future electricity bill and the future failure handling cost is smaller , With time determination function,
The expense prediction apparatus according to claim 4. The cost prediction apparatus according to any one of claims 1 to 5, wherein the failure-corresponding cost prediction function calculates a repair cost, a replacement cost, or both based on a variable or a constant set for each facility. .   The program for functioning a computer as a cost prediction apparatus as described in any one of Claims 1-6.

Images