Utility diagnosing equipment, operational program therefor, and utility diagnosing method

• US 7,424,394 B2
• Filed: 08/26/2005
• Issued: 09/09/2008
• Est. Priority Date: 08/27/2004
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Suzuki, Hideaki, Yokota, Toshimi, Saito, Masahiko, Taira, Tomotsune
• Primary Examiner(s): Barlow, Jr.; John E
• Assistant Examiner(s): Washburn; Douglas N
• Application Number:
  US11/211,434
• Publication Number:
  US 20060136175A1
• Time in Patent Office: 1,110 Days
• Field of Search: 702/182, 702/183
• US Class Current: 702/182
• CPC Class Codes: F24F 11/30 for purposes related to the operation of the system, e.g. for safety or monitoring
• CPC Class Codes: F24F 11/38 Failure diagnosis
• CPC Class Codes: F24F 11/63 Electronic processing
• CPC Class Codes: G05B 2219/2638 Airconditioning
• CPC Class Codes: G05B 23/0235 based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold

Abstract

A utility diagnosing equipment can recognize an abnormality or a normality in a plurality of utilities of an identical type, which would not be recognized when an absolute threshold is set. The equipment accepts state values from the plurality of utilities, sets an average value of the state values of each utility to a proxy value, finds a relative value between the state value of each utility and the proxy value thereof, and displays the relative value for each utility. The relative values of the utilities exceeding a predetermined threshold is highlighted.

Claims

1. A utility diagnosing equipment for diagnosing a plurality of utilities of an identical type on the basis of state values from the utilities, comprising:

state value accepting means for accepting said state values from said plurality of utilities;

state value memory means for storing, for each utility, a time-series state value of each utility;

completing means for estimating a missing state value in the time-series state value of each utility stored in said state value memory means;

proxy value calculating means for taking statistics of the state values from said respective utilities to calculate a proxy value representative of states of each utility;

relative value calculating means for finding a relative value for each utility between the state value of each utility and said proxy value; and

output means for outputting said relative value for each utility and/or utility identification information about the utility when said relative value exceeds a predetermined range;

wherein said proxy value calculating means calculates a completed proxy value at each time as said proxy value with use of the state value at each time estimated by said completing means, and said output means outputs said relative value for each utility and/or utility identification information about the utility when said relative value exceeds a predetermined range; and

wherein said completing means finds a moving average of the state values of the utilities at the respective times as a moving average width of a predetermined time width and completes the missing state value.

2. A utility diagnosing equipment according to claim 1, wherein said proxy value calculating means uses the average value of the state values of the utilities at the respective times completed by said completing means as said completed proxy value at each time.

3. A utility diagnosing system comprising:

utility diagnosing equipment set forth in claim 1; and

an information collecting device for collecting said state values from the utilities and transmitting the state values to said state value accepting means of said utility diagnosing equipment.

4. A computer-readable medium having embedded therein, a utility diagnosing program for diagnosing a plurality of utilities of an identical type on the basis of state values from the utilities, wherein said program causes a processor to execute:

a state value accepting step of accepting said state values from said plurality of utilities;

a state value memory step of storing time-series state values for each utility accepted in said state value accepting step;

a completing step of completing a missing state value in the time-series state values of the utility stored in said state value memory step;

a proxy value calculating step of taking statistics of the state values from said plurality of utilities to calculate a proxy value representative of the states of each utility;

a relative value calculating step of finding a relative value for each utility between the state value of each utility and said proxy value; and

an output step of outputting said relative value for each utility and/or utility identification information about the utility when said relative value exceeds a predetermined range;

wherein a completed proxy value at each time is calculated as said proxy value with use of the state value of the utilities at the time completed in said completing step in said proxy value calculating step, and a relative value between the state value of each utility at each time completed in said completing step and said completed proxy value of each time and/or utility identification information about the utility when the relative value exceeds a predetermined range, are/is output in said output step; and

wherein a moving average of the state values of the utilities at the respective times is found as a moving average width of a predetermined time width to complete the missing state value in said completing step.

5. A computer-readable medium according to claim 4, an average value of the state values of the utilities at the respective times completed in said completing step is used as said completed proxy value at each time in said proxy value calculating step.

6. A computer-readable medium according to claim 4, wherein said program causes a processor to execute:

an allowable range accepting step of accepting said predetermined range for said relative value; and

an abnormality decision step of deciding whether or not said relative value of each utility at each time is in said predetermined range accepted in said allowable accepting step,

wherein, in said output step, when it is decided in said abnormality decision step that there is a relative value not in said predetermined range among said relative values of the utilities at the respective times, the relative value is highlighted.

7. A computer-readable medium according to claim 6, wherein a processor executes a self diagnosis accepting step of accepting an abnormality notification indicating that said state value issued from at least one of the plurality of utilities is abnormal and an allowable range setting step of, when said abnormality notification is accepted in said self diagnosis accepting step, determining said predetermined range on the basis of said relative value of the utility determined as abnormal at an abnormality detection time and passing the predetermined range to said allowable range accepting step.

8. A computer-readable medium according to claim 4, wherein a processor executes a sampling density calculating step of finding a sampling density from a sampling number of said state value with the predetermined time width for each utility, and said sampling number is output together with said relative value at each time in said output steps.

9. A computer-readable medium according to claim 4, wherein a graph including a relative value axis having said relative value as a parameter, a time axis having time as a parameter, and a utility axis having a utility identification information as a parameter and showing the relative value at each time for each utility is output in said output step.

10. A computer-readable medium according to claim 9, wherein, when there is one of the time-series relative values of the utilities exceeding a predetermined range with said time passage, the relative value in said graph is highlighted in said output step.

11. A computer-readable medium according to claim 9, wherein said completed proxy value at each time for each utility is displayed in said graph and the relative value at each time for each utility is displayed with the completed proxy value at each time displayed for each utility as a reference in said output step.

12. A computer-readable medium according to claim 4, wherein, when there are a plurality of types of state values as the state values of said utilities, a processor executes said state value accepting step and said proxy value calculating step for each state value, and

a graph including a relative value axis having said relative value as a parameter, a time axis having time as a parameter, and a state value type axis having the type of each state value as a parameter and indicating the relative value at each time for each type of each state value with regard to one of the plurality of utilities is output in response to an external instruction in said output step.

13. A computer-readable medium according to claim 4, wherein a processor executes a relative value memory step of storing said relative value found in said relative value calculating step, and

a graph including a relative value axis having said relative value as a parameter, a period-by-period axis having period as a parameter, and a time axis having time in each period as a parameter and indicating the relative value at each time in each period is output in response to an external instruction.

14. A utility diagnosing method for diagnosing a plurality of utilities of an identical type on the basis of state values from the plurality of utilities, comprising:

a state value accepting step of accepting said state values from the plurality of utilities;

a state value memory step of storing time-series state values for each utility accepted in said state value accepting step;

a completing step of completing a missing state value in the time-series state values of the utilities stored in said state value memory step;

a proxy value calculating step of taking statistics of the state values from the plurality of utilities to calculate a proxy value representative of the states of each utility;

a relative value calculating step of a relative value between the state value of each utility and said proxy value; and

an output step of outputting said relative value for each utility and/or utility identification information on the utility when said relative value exceeds a predetermined range;

wherein a completed proxy value at each time as said proxy value is calculated in said proxy value calculating step with use of the state value at each time completed in said completing step, and in said output step, a relative value between the state value at each time completed in said completing step and said completed proxy value at each time is output and/or utility identification information on the utility is output when said relative value exceeds a predetermined range; and

wherein said completing step finds a moving average of the state values of the utilities at the respective times as a moving average width of a predetermined time width and completes the missing state value.