JP2018185696A - Deterioration diagnostic system and deterioration diagnostic method of air conditioner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide, in a simple configuration and an inexpensive manner, a deterioration diagnostic system and a deterioration diagnostic method, capable of determining deterioration of an air conditioner mounted on a vehicle.SOLUTION: A deterioration diagnostic system comprises: an air conditioner AC which controls a temperature in a vehicle room; indoor temperature measuring means Ti for measuring data at an outdoor air temperature Te and an indoor temperature of a vehicle; indoor temperature predicting means PM for predicting an indoor temperature Ti from the outdoor air temperature Te; and deterioration determining means DM for determining the deterioration of the air conditioner AC on the basis of a difference ΔT between the indoor temperature Ti and a prediction temperature Tp obtained by the indoor temperature predicting means PM. The deterioration determining means DM determines the deterioration of the air conditioner AC on the basis of probability distribution of a prediction error ΔT obtained by subtracting the prediction temperature Tp from the indoor temperature Ti.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a deterioration diagnosis system and a diagnosis method for detecting and determining a deterioration in performance of an air conditioner over time, and in particular, a deterioration diagnosis system and a diagnosis method suitable for an air conditioner mounted on a moving body such as a railway vehicle. It is about.

  Conventionally, air conditioners mounted on moving bodies such as railway vehicles used by an unspecified number of passengers are not limited to regular maintenance inspections of vehicles, but maintenance inspections are performed separately from the periodic inspections. Has been done. This is because a decrease in the performance of the air conditioner is directly linked to a deterioration in the interior environment of the vehicle, and coupled with recent warming, the decrease in the cooling performance of the air conditioner particularly in the summertime is a great deal for passengers. This is to cause discomfort. Thus, it is strongly desired to implement timely degradation diagnosis of such an air conditioner, and various diagnostic systems have been proposed (see, for example, Patent Documents 1 to 4).

  Here, Patent Document 1 discloses an air conditioning management system that grasps changes in the temperature inside the vehicle from various environmental information such as outside air temperature, inside humidity, and boarding rate, and actual environment information, and determines the deterioration state of the air conditioner. In addition, an air conditioning management method is disclosed.

  Patent Document 2 discloses a method for determining deterioration of an air conditioner based on an accumulated operation time for each air conditioner. Patent Document 3 discloses a method for determining deterioration from a comparison between an operation pattern of a compressor and a reference pattern. Patent Document 4 discloses a method for determining deterioration based on a change in power consumption.

Reissue 2009/150724 JP 2014-69630 PR JP 2007-83872 A JP 2009-20721 A

  However, any of the prior arts disclosed in Patent Documents 1 to 4 described above includes means for acquiring various environmental information (for example, a variable load sensor, a temperature / humidity sensor, etc. for detecting the weight of the vehicle in order to acquire the boarding rate). ) And a variety of data analysis processing, etc., and a communication system for transmitting information is required, making the system complex and large-scale and economical only on routes with many passengers in the metropolitan area and urban areas. There were concerns that it would be difficult to hire them.

  Moreover, in the prior art disclosed in Patent Document 1, there has been a problem that a temporary temperature change caused by a disturbance such as weather or door opening / closing is erroneously determined as deterioration of the air conditioner.

  Similarly, in the prior arts disclosed in Patent Documents 2 and 3, there is a problem that the deterioration cannot be correctly determined if the deterioration state differs for each device even if the operation operation time and the integrated operation time are the same. In the prior art, the power supply amount changes depending on the travel location and travel state of the railway vehicle, so that the power consumption also changes, and there is a problem that the deterioration cannot be correctly determined.

  Accordingly, the present invention has been made in view of the above-described problems of the prior art, and a deterioration diagnosis system and a deterioration diagnosis method capable of determining deterioration of an air conditioner mounted on a vehicle with a simple configuration. The purpose is to provide it at low cost.

  In order to achieve the above object, a deterioration diagnosis system for an air conditioner according to the present invention includes an air conditioner that adjusts the temperature of a vehicle interior, outside air temperature data, an indoor temperature measuring means that measures an indoor temperature of the vehicle, Room temperature prediction means for predicting the room temperature from outside temperature, and deterioration determination means for judging deterioration of the air conditioner based on a difference between the room temperature and a predicted temperature obtained from the room temperature prediction means. The deterioration determination means performs the deterioration determination of the air conditioner based on a probability distribution of a prediction error obtained by subtracting the predicted temperature from the room temperature.

  In the present invention, the deterioration of the air conditioner means a deterioration of performance over time (decrease in capacity). Specifically, the cooling performance of the air conditioner is deteriorated over time due to contamination or clogging of the radiator or the fan. It shall mean the lowered state.

  When configured in this way, deterioration diagnosis of an air conditioner mounted on a vehicle can be performed by simple data processing based on the minimum necessary data.

  Further, the deterioration determining means calculates a deterioration degree based on the probability distribution of the prediction error, and the deterioration degree of the air conditioner is determined when a cumulative deterioration degree obtained by accumulating the deterioration degree over time exceeds a predetermined value. You may judge.

  In the case of such a configuration, it is possible to prevent an erroneous determination due to a sudden abnormal value by performing the deterioration determination based on the cumulative deterioration degree in which the prediction errors are accumulated in time series.

Further, when the degree of deterioration is a, the prediction error is ΔT, the probability density is P (ΔT), and C is a constant, the degree of deterioration a can be expressed by Equation (1).
a = −log (P (ΔT)) + C Expression (1)

  When configured in this way, it is possible to simply and quantitatively calculate the deterioration degree a that is the basis for the deterioration determination.

Further, when the deterioration degree is a, the prediction error is ΔT, the probability density is P (ΔT), and C is a constant, the deterioration degree a can be expressed by Expression (2).
a = −log (P (ΔT)) + C (ΔT> 0)
a = −log (P (0)) + C (ΔT ≦ 0) (2)

  In the case of such a configuration, it is possible to simply and specifically simply calculate the deterioration degree a with higher accuracy according to the region where the prediction error ΔT exists.

  Further, the constant C may be set so that the minimum value of the deterioration degree a is 0 or less and the maximum value is positive within a predetermined measurement period.

  When configured in this way, the degree of deterioration a can be changed positively or negatively based on the region where the prediction error ΔT exists (the cumulative degree of deterioration can be increased or decreased), and the degree of deterioration with higher accuracy according to the state of the air conditioner It can be simply calculated specifically and quantitatively.

Furthermore, _{when S t} a cumulative deterioration degree at time t, the deterioration degree at time t _{a t,} a cumulative deterioration degree at time t-1 and the _{S t-1,} the accumulated deterioration degree _{S t} has the formula (3) Can be expressed as
S _t = S _t−1 + a _t (3)

  When configured in this way, the deterioration over time (cumulative deterioration) accumulated over time can be used to more accurately grasp the changes over time of the air conditioner, and the cumulative deterioration can be determined concretely and quantitatively. Simple calculation.

Furthermore, _{when S t} a cumulative deterioration degree at time t, the deterioration degree at time t _{a t,} a cumulative deterioration degree at time t-1 and the _{S t-1,} the accumulated deterioration degree _{S t} has the formula (4) Can be expressed as
S _t = max (S _t−1 + a _t , 0) (4)

  Here, max (◯, ○) in the above formula (4) means the larger of the two values in ().

When configured in this way, it is possible to more accurately grasp changes over time of the air conditioner using the deterioration degree accumulated over time (cumulative deterioration degree), and the normal state of the air conditioner continues. (when S _t-1 + a _t is 0 or less), to prevent the cumulative degradation degree S _t continues to decrease (reset cumulative deterioration degree S _t), the detection of outliers (deterioration degree plus) Sensitivity can be increased.

  In the above, the probability distribution may be a normal distribution.

  When configured in this manner, a probability distribution model that is the basis for the deterioration determination can be easily set.

  The deterioration diagnosis method for an air conditioner according to the present invention measures an indoor temperature using an air conditioner that regulates the temperature of the vehicle interior, outside air temperature data, and a regression equation between the outside air temperature and the room temperature that are obtained in advance. , Using the regression equation to calculate the predicted temperature from the outside air temperature data, calculate the degree of deterioration based on the probability density obtained by applying the difference between the measured indoor temperature and the calculated predicted temperature to the normal distribution, When the cumulative deterioration level accumulated in time series exceeds a predetermined threshold value, it is determined that the air conditioner has deteriorated.

  In such a configuration, based on the minimum necessary data, by performing deterioration determination based on the deterioration degree accumulated in time series by simple data processing, erroneous determination due to sudden abnormal values can be made in advance. Can be prevented.

  In addition, the cumulative deterioration degree is set to be adjustable within a predetermined measurement period, and when the cumulative deterioration degree is positive, the cumulative deterioration degree is accumulated, while the cumulative deterioration degree is zero or negative. In this case, the accumulated cumulative deterioration level may be reset.

  When configured in this way, it is possible to more accurately grasp the change over time of the air conditioner using the cumulative deterioration degree set so as to be able to increase or decrease, and when the normal state of the air conditioner continues (cumulative deterioration degree) If the value is less than or equal to 0), the cumulative deterioration degree is temporarily reset (set to 0) to prevent the cumulative deterioration degree from continuing to decrease, and the detection sensitivity of subsequent abnormal values (deterioration degree plus) is increased. Can be increased.

  According to the present invention, it is possible to perform deterioration diagnosis of an air conditioner mounted on a vehicle by simple data processing based on the minimum necessary data without requiring analysis processing of various kinds of data.

It is a functional block diagram which shows the structure of the deterioration diagnosis system of the air conditioner which concerns on this invention. It is a mimetic diagram showing probability distribution model P in one embodiment of an air-conditioner degradation diagnostic system concerning the present invention. It is a schematic diagram which shows the relationship between prediction error (DELTA) T and the deterioration degree a in one Embodiment of the deterioration diagnosis system of the air conditioner which concerns on this invention. It is a schematic diagram which shows the relationship between the probability distribution model P and the deterioration degree a in one Embodiment of the deterioration diagnosis system of the air conditioner which concerns on this invention.

  An embodiment of a deterioration diagnosis system for an air conditioner according to the present invention will be described below with reference to the drawings. Here, FIG. 1 is a functional block diagram showing the configuration of the deterioration diagnosis system for an air conditioner according to the present invention, and FIGS. 2 to 4 are for explaining the relationship between the probability density distribution and the degree of deterioration in the present embodiment. It is a schematic diagram.

  As schematically shown in FIG. 1, the deterioration diagnosis system 1 for an air conditioner according to the present embodiment includes, for example, an air conditioner AC that air-conditions the interior of a railway vehicle, and an indoor temperature measuring means that measures an indoor temperature Ti. A room temperature predicting means PM for predicting the room temperature Ti by regression analysis based on the correlation between the outside air temperature Te and the room temperature Ti, a room temperature measured value Ti, and a room temperature predicted value Tp. Deterioration determining means DM for determining the deterioration of the air conditioner AC based on the difference ΔT (the measured indoor temperature value Ti−the predicted indoor temperature value Tp).

  In the present embodiment, external public data such as the Japan Meteorological Agency can be used as the outside air temperature Te. Of course, the outside air temperature Te may be measured by a measuring means such as an external thermometer separately provided on the vehicle or the ground.

  Further, as the room temperature Ti, for example, data measured by a room thermometer provided at a return port (suction port for sucking indoor air) of the air conditioner AC can be used.

  Next, the procedure of deterioration diagnosis when the deterioration diagnosis system and the deterioration diagnosis method according to the present embodiment are applied to an air conditioner mounted on a railway vehicle will be described.

(Preparation)
In the deterioration diagnosis system and diagnosis method according to the present embodiment, first, data on the room temperature Ti and the outside air temperature Te are collected in advance in a railway vehicle equipped with the same type of air conditioner AC.

  Then, linear regression analysis is performed based on the collected data of the indoor temperature Ti and the outside air temperature Te, and a correlation between the room temperature Ti and the outside air temperature Te is obtained. Specifically, a regression equation from the room temperature Ti to the outside air temperature Te is created (for example, the room temperature Ti = β1 × the outside air temperature Te + β2). Here, any regression equation can be used as long as the correlation between the outside air temperature Te and the room temperature Ti can be defined (the room temperature Ti can be predicted from the outside air temperature Te). It is also possible to use a multiple regression analysis with explanatory variables other than the outside air temperature added. The correlation formula (regression formula) corresponds to the room temperature prediction means PM of the present invention.

Next, a probability distribution model P (ΔT) of a residual (hereinafter also referred to as a prediction error) ΔT obtained by subtracting the predicted indoor temperature value Tp from the measured indoor temperature value Ti is created. The probability distribution model is approximated as a normal distribution model N (0, σ ² ) having an average value μ = 0, and a standard deviation σ is calculated. However, the average value μ and the standard deviation σ of the normal distribution model may be other values, and a mixed Gaussian distribution model or an arbitrary probability distribution model may be used.

(Judgment)
The indoor temperature prediction means PM according to the present embodiment calculates the predicted indoor temperature value Tp from the newly acquired outside air temperature Te using the regression equation. Then, a prediction error ΔT obtained by subtracting the predicted indoor temperature value Tp from the measured indoor temperature value Ti is calculated.

  Next, the degradation determination means DM according to the present embodiment applies the calculated prediction error ΔT to the probability distribution model, and calculates the probability density P (ΔT).

  Next, based on the probability density P (ΔT), the deterioration degree a is calculated by the following equation.

  a = −log (P (ΔT)) + C (when ΔT> 0)

  a = −log (P (0)) + C (when ΔT ≦ 0)

The constant C is the following equation using the probability distribution model P (ΔT) and the standard deviation σ.
C = log (P (σ))

  Here, the reason why the constant C is a function of P (σ) is to enable statistical processing based on the same standard regardless of variations in the air conditioner and the vehicle.

  In the present embodiment, by setting the constant C as described above, the deterioration degree a is calculated using + 1σ as a reference value as schematically shown in FIG. That is, as schematically shown in FIGS. 3 and 4, when the prediction error ΔT (the measured indoor temperature value Ti−the predicted indoor temperature value Tp) is larger than the standard deviation + 1σ, the degree of deterioration a is positive, and the prediction error ΔT is When the standard deviation is + 1σ or less, the deterioration degree a is set to be 0 or less.

  In the present embodiment, the standard deviation 1σ on the + side is set as the reference value of the deterioration degree a because the prediction error ΔT (substantially greater than + 1σ, which is considered not to occur continuously if normal) 16%) is understood as a performance deterioration (insufficient cooling) due to aged deterioration of the air conditioner AC.

  The constant C is not limited to the above formula, but it is preferable to set the minimum value of the deterioration degree a in the predetermined measurement period MT to be 0 or less and the maximum value to be positive. In other words, it is preferable to set the constant C so that the cumulative deterioration degree St can be increased or decreased within a predetermined measurement period MT (so that the deterioration degree a can be changed over the positive and negative regions). . As a result, even if the cumulative deterioration St increases temporarily due to disturbances such as weather or door opening / closing, if the normal state (deterioration a is negative) continues thereafter, the cumulative deterioration St decreases. Therefore, it is possible to prevent a situation where the deterioration is erroneously determined.

  The reference value for reversing the sign of the deterioration degree a (increasing or decreasing the cumulative deterioration degree St) is basically preferably set to an integral multiple of the standard deviation σ. However, for example, when + 2σ is adopted as the reference value, the sensitivity is significantly dull (degradation degree a corresponding to approximately 98% of the prediction error ΔT is set to a negative value), and in most cases, the deterioration degree a is negative. The plus value of the cumulative deterioration degree St described later will not be accumulated. Therefore, in practice, it is more preferable to set the reference value to about + 1σ as schematically shown in FIGS. Thereby, a prediction error ΔT of + 1σ or less (approximately 84%) is detected as a normal value (deterioration degree minus), and a prediction error ΔT (approximately 16%) exceeding + 1σ is detected as an abnormal value (deterioration degree plus). Can do.

  Further, as best shown in FIG. 3, the deterioration degree a is a fixed minimum value (in this example, the minimum when the average value μ = 0 of the normal distribution) in a region where the prediction error ΔT is 0 or less. Is preferred.

  In this way, in the negative region of the prediction error ΔT (region where the room is sufficiently cooled), the deterioration degree a is set to a constant value (minimum value), so that sudden abnormal values (for example, due to manual operation) It is possible to prevent an erroneous determination at the time of deterioration diagnosis by eliminating the influence of temporary supercooling during forced cooling operation or the like.

  From the viewpoint of determining deterioration based on temperature data that can approximate a normal distribution, the measurement period MT is provided with a measurement period of about 1 to 2 weeks with the sampling period of the room temperature Ti being about 1 minute. This is possible.

Then, deterioration determining unit DM according to the present embodiment, based on the probability distribution, the cumulative degree of deterioration S _t-1 at time t-1 to cumulative degradation level S _t at time _t, using the degree of deterioration a _t It is calculated by the following formula. In the following expression, max (◯, ○) means the larger of the two values in ().

S _t = max (S _t−1 + a _t , 0)

That is, when S _t-1 + _{a t} is larger than 0, the value of the cumulative degradation degree _{S t,} is set to S _t-1 + _{a t.} On the other hand, if S _t-1 + _{a t} is 0 or less, (resets) the value of the cumulative degree of deterioration _{S t} is set to 0.

In this way, by using the cumulative deterioration degree St obtained by accumulating the deterioration degree a in time series, it becomes possible to grasp the change over time of the air conditioner more accurately. Further, when a normal state of the air conditioner AC has been continued (if S _t-1 + a _t is 0 or less), to prevent the cumulative degradation degree S _t continues to decrease (reset cumulative deterioration degree S _t) Thus, the detection sensitivity of abnormal values (deterioration degree plus) can be increased.

When the cumulative degradation level S _t exceeds a predetermined threshold value, to perform maintenance of the air conditioner. Note that the above-described normal / abnormal deterioration determination of the air conditioner may be performed in each railway vehicle, or may be centrally managed using a communication line in a facility such as a station or a garage. Further, the determination may be made in real time, or the batch method may be used in which data is collected at a certain period and then collectively determined.

  Further, regarding the threshold value, for example, the durability of the air conditioner such as setting the value of the cumulative deterioration degree St when the state of the deterioration degree a of about 1.5σ continues for several days (for example, 3 days) is set as the threshold value. It can be set as appropriate according to usage.

  Also, the threshold value is set stepwise (for example, set to the value of the cumulative deterioration St when the prediction error ΔT is a value of 1.3σ, 1.5σ, 2σ, etc. for three consecutive days), and according to each step Then, subsequent maintenance processing may be set in stages. Specifically, for example, the threshold level 1 is set with a prediction error of 1.3 σ and the cumulative deterioration degree value corresponding to three consecutive days is set as a low level of urgency requiring attention, and the threshold level is set with a prediction error of 1.5 σ. The cumulative deterioration value corresponding to 3 days is set as medium level of emergency requiring maintenance within a few weeks, and the cumulative degradation level corresponding to 3 consecutive days with a prediction error of 2σ is set as the threshold level 3 for emergency requiring emergency maintenance. It can be set in steps such as degree.

  As described above, according to the deterioration diagnosis system and deterioration diagnosis method for an air conditioner according to the present invention, publicly available data can be used for the outside air temperature Te, and the measured data is air-conditioning only with the room temperature Ti in the vehicle. Since the deterioration diagnosis of the apparatus AC can be performed, a very simple configuration can be achieved, which can be easily realized as a software program and can contribute to a significant cost reduction. Further, since a large-scale communication system is not required, for example, it can be individually installed in each vehicle such as a personal computer in which the components of the deterioration diagnosis system (indoor temperature prediction means PM, deterioration determination means DM, etc.) are installed as software. It can be easily configured as a simple control device.

  The technical scope of the present invention is not limited to the above-described embodiments, and various changes or improvements can be added without departing from the scope of the present invention. For example, the deterioration diagnosis system for an air conditioner according to the present invention can be easily applied not only as a railway vehicle but also as a deterioration diagnosis system for an air conditioner mounted on another moving body such as an automobile or an aircraft.

1: Deterioration diagnosis system a: Deterioration degree AC: Air conditioner C: Constant DM: Degradation determination means MT: Measurement period PM: Indoor temperature prediction means _St : Cumulative deterioration degree Te: Outdoor air temperature Ti: Indoor temperature measurement value Tp: Indoor Predicted temperature ΔT: Prediction error μ: Average value σ: Standard deviation

Claims (10)

An air conditioner for controlling the temperature of the vehicle interior;
Outside temperature data,
Indoor temperature measuring means for measuring the indoor temperature of the vehicle;
Indoor temperature predicting means for predicting the indoor temperature from the outside air temperature;
Deterioration determining means for determining deterioration of the air conditioner based on the difference between the indoor temperature and the predicted temperature obtained by the indoor temperature predicting means;
The deterioration diagnosis unit for an air conditioner, wherein the deterioration determination unit performs the deterioration determination of the air conditioner based on a probability distribution of a prediction error obtained by subtracting the predicted temperature from the room temperature.   The deterioration determining means calculates a deterioration degree based on the probability distribution of the prediction error, and determines that the air conditioner has deteriorated when a cumulative deterioration degree accumulated over time exceeds a predetermined value. The deterioration diagnosis system for an air conditioner according to claim 1. 3. The deterioration degree a can be expressed by Equation (1), where a is the deterioration degree, ΔT is the prediction error, P (ΔT) is the probability density, and C is a constant. Air conditioner deterioration diagnosis system.
a = −log (P (ΔT)) + C Expression (1) 3. The deterioration degree a can be expressed by equation (2), where a is the deterioration degree, ΔT is the prediction error, P (ΔT) is the probability density, and C is a constant. Air conditioner deterioration diagnosis system.
a = −log (P (ΔT)) + C (ΔT> 0)
a = −log (P (0)) + C (ΔT ≦ 0) (2)   5. The air conditioning according to claim 3, wherein the constant C is set so that the minimum value of the deterioration degree a is 0 or less and the maximum value is positive within a predetermined measurement period. Equipment deterioration diagnosis system. When the cumulative degradation level at time t is S _t , the degradation level at time t is a _t , and the cumulative degradation level at time t−1 is S _t−1 ,
6. The deterioration diagnosis system for an air conditioner according to claim 2, wherein the cumulative deterioration degree _St is expressed by an expression (3).
S _t = S _t−1 + a _t (3) When the cumulative degradation level at time t is S _t , the degradation level at time t is a _t , and the cumulative degradation level at time t−1 is S _t−1 ,
6. The deterioration diagnosis system for an air conditioner according to claim 2, wherein the cumulative deterioration degree _St is expressed by an expression (4).
S _t = max (S _t−1 + a _t , 0) (4)   The deterioration diagnosis system for an air conditioner according to any one of claims 1 to 7, wherein the probability distribution is a normal distribution. Using an air conditioner that regulates the temperature of the vehicle interior, outside air temperature data, and a regression equation between the outside air temperature and the room temperature determined in advance,
The indoor temperature is measured, the indoor temperature is calculated from the outside air temperature data using the regression equation, and the degree of deterioration based on the probability density obtained by applying the difference between the measured indoor temperature and the calculated predicted temperature to the normal distribution is calculated. Calculate
A deterioration diagnosis method for an air conditioner, wherein the air conditioner is determined to have deteriorated when an accumulated deterioration degree in which the deterioration degree is accumulated in a time series exceeds a predetermined threshold. When the cumulative deterioration degree is set to be adjustable within a predetermined measurement period, and when the cumulative deterioration degree is positive, the cumulative deterioration degree is accumulated, while when the cumulative deterioration degree is zero or negative. 10. The method for diagnosing deterioration of an air conditioner according to claim 9, wherein the accumulated deterioration degree is reset.

Images