Health status evaluation system of hydropower units based on multi-source information fusion

Abstract

The present invention discloses a hydropower unit health status evaluation system based on multi-source information fusion, which relates to the technical field of hydropower unit health status evaluation. A stage analysis unit is used to determine whether the same type of equipment will be in any of the basic period, the two-term period and the final period under different service times, and then the content of the enabled monitoring object is analyzed according to the different time periods; the monitoring object analyzes several maintenance data, observes the number of occurrences of each monitoring data under abnormal conditions, and determines which monitoring objects are in high frequency according to the number of occurrences, so as to determine to monitor different data at different stages; thereby, the present application can observe a small amount of data under different conditions to determine whether an abnormality occurs. The present application is simple and effective, does not need to monitor all data, and can save computing power in some cases.

Description

Technical Field

[0001] The present invention belongs to the technical field of health evaluation of hydropower units, and in particular is a health status evaluation system of hydropower units based on multi-source information fusion. Background Art

[0002] As the core equipment of hydropower stations, the health status of hydropower units is directly related to the stability and safety of power supply. At present, some progress has been made in the research on health assessment of hydropower units. Traditional methods mainly rely on regular offline detection and testing to monitor the vibration, temperature, pressure and other physical quantities of the unit to determine the health status of the equipment. Although these methods can detect potential faults to a certain extent, they have problems such as long detection cycle and delayed fault detection.

[0003] In recent years, with the development of sensor technology and big data technology, more and more research has begun to focus on health assessment methods based on multi-source information fusion. For example, a combination of vibration analysis, oil analysis, acoustic emission technology and other methods can be used to achieve comprehensive monitoring and evaluation of the unit's operating status. In addition, some studies have introduced machine learning and artificial intelligence algorithms to automatically identify and predict the unit's health status by analyzing and mining a large amount of monitoring data.

[0004] However, the health assessment of hydropower units generally adopts full data monitoring. This process does not take into account the service life of the hydropower units, resulting in a waste of monitoring computing power. Based on this, a solution is provided. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art;

[0006] To this end, the present invention proposes a hydropower unit health status evaluation system based on multi-source information fusion, including:

[0007] The acquisition control unit is used to synchronize the monitored real-time object, determine its real-time period according to its working time, and perform status evaluation operations according to the real-time period. The specific operation method is as follows:

[0008] When the real-time period is in the basic period, the monitoring unit will be driven to obtain the real-time value of the data corresponding to the basic data name, and transmit the real-time data to the multi-source analysis unit to analyze whether the real-time object is abnormal;

[0009] If the real-time period is in the binomial period, the monitoring unit will be driven to obtain the real-time values ​​of the data corresponding to the basic data name and the binomial data name, and the real-time data will be transmitted to the multi-source analysis unit to analyze whether the real-time object is abnormal;

[0010] If the real-time period is in the last period, the monitoring unit will be driven to obtain the real-time values ​​of all data corresponding to the basic data name, the second data name and the last data name, and the real-time data will be transmitted to the multi-source analysis unit to analyze whether the real-time object is abnormal;

[0011] When the multi-source analysis unit detects an abnormality, it will transmit an abnormal signal to the display unit, and the display unit will automatically display the abnormal signal when it receives it.

[0012] Furthermore, the real-time period is realized by performing stage summary analysis through the stage determination unit. The specific method of stage analysis is as follows:

[0013] Select any equipment of the same type as any equipment in the hydropower unit that has been scrapped in the past six months and mark it as the target object;

[0014] The total service time and maintenance times of the target object are obtained. Here, the average total service time of all scrapped equipment is used to determine the maintenance times. Similarly, the maintenance times are also determined according to the average maintenance times of several scrapped hydropower units.

[0015] First, the total service time is preliminarily divided into three stages, namely the basic period, the two-term period and the final period. Then, the number of maintenance times and the maintenance time of the corresponding equipment in each stage are calculated. The maintenance time is divided by the number of maintenance times to determine the average maintenance time. Then, according to the difference between the average maintenance times of two adjacent maintenance segments, that is, the segment difference, and the relationship between the segment difference and the set values ​​X1 and X2, the duration of the basic period, the two-term period and the final period is adjusted.

[0016] After adjustment, determine the basic period, binomial period and final period.

[0017] Furthermore, the specific method for determining the basic period, the binomial period and the final period is as follows:

[0018] Set a basic time period and mark it as T1. T1 is a preset value. The value of T1 generally does not exceed one third of the total service time.

[0019] Then, the number of times the corresponding target object is repaired in the basic period is obtained, which is marked as the maintenance segment number; and the total maintenance time is marked as the maintenance segment time, and the maintenance segment time is divided by the maintenance segment number to obtain a value marked as the average maintenance segment time;

[0020] Then, the second third time period after the basic time period is obtained, that is, the time period corresponding to one third of the total service time after the basic time period, and it is marked as the binomial time period, and the rest is marked as the final time period; the maintenance time and maintenance times of the binomial time period and the final time period are obtained simultaneously;

[0021] Here, the maintenance time and maintenance times of the basic period, the two-term period and the final period are calculated by the average values ​​of several target objects of the same type in the corresponding different periods. Here, if a specific example is given, it means that one target object is selected, and then the maintenance times and maintenance time of several target objects of the same type in the basic period are obtained, and then the average values ​​of several maintenance times are calculated to represent the maintenance times of the target object;

[0022] The dimension segment average time of the second period and the final period is obtained, and marked as dimension segment average time 2 and dimension segment average time 3 respectively;

[0023] Then, the difference between the mean time of the second dimension segment and the mean time of the dimension segment is obtained, marked as the segment difference value, and the basic period and the binomial period are adjusted according to the segment difference value;

[0024] Update the binomial period. When the basic period and binomial period are updated, the remaining period is marked as the last period.

[0025] Get the basic period, binomial period and final period.

[0026] Furthermore, the specific method of adjusting the basic period and the binomial period according to the segment difference value is as follows:

[0027] When X1≤segment difference≤X2, no processing is done at this time. X1 and X2 are both preset values, indicating that the division of the basic period and the binomial period is reasonable. If the segment difference is less than X1, the basic period will be automatically reduced by the set duration T2 in sequence. The reduction method is to push the end time point of the basic period forward by T2, and push the start time point of the binomial period forward by T2. The basic period and the binomial period are updated, and the segment difference is recalculated after the update. If the segment difference is still less than X1, the T2 duration will continue to be reduced, and the basic period and the binomial period will be updated until the segment difference satisfies X1≤segment difference≤X2;

[0028] If the segment difference is greater than X2, the end time point of the binomial period will be pushed forward by T2, and then the segment difference between the updated binomial period and the basic period will be recalculated. If it is still greater than X2, it will continue to be pushed forward by T2 until the segment difference satisfies X1≤segment difference≤X2.

[0029] Furthermore, it also includes a data collection unit for collecting data on the same hydropower units that have been scrapped in the past six months, and obtaining periodic information of the hydropower units. The periodic information specifically includes: maintenance times, the duration of each maintenance, and the total service duration. The maintenance times here refer to the number of times the equipment of the corresponding hydropower unit has been repaired before being scrapped, and the total service duration refers to the total usage time of the corresponding equipment. It is based on observing the total service duration of several identical hydropower units after being scrapped and then taking the average value. The maintenance times are also determined by the average of the equipment of several hydropower units.

[0030] Furthermore, the data collection unit is also used to collect each maintenance information of the corresponding target object reported for reimbursement in the past six months. The maintenance information includes the abnormal data name and abnormal duration. The abnormal data name is the name of the element that deviates when the corresponding abnormality occurs, such as stator temperature, rotor temperature, etc. The abnormal duration is the time required for maintenance and recovery after the corresponding abnormality occurs, and the abnormal data name and abnormal duration are transmitted to the stage determination unit. The stage determination unit is used to analyze the abnormal data name and abnormal duration for abnormal labels. The specific method of abnormal label analysis is as follows:

[0031] Select any maintenance information, assign the abnormal duration to the abnormal data name, and re-mark it as an abnormal value. Add up the abnormal values ​​of all the same abnormal data names, and mark the obtained value as the abnormal total value, so as to obtain the abnormal total value of each abnormal data name;

[0032] Get the number of times the abnormal data name appears, mark it as abnormal times, and use the formula to calculate the abnormal recognition value, abnormal recognition value = 0.62 × abnormal times + 0.38 × total abnormal value;

[0033] Sort the abnormal data names according to the total abnormal value, and determine the basic data name, binomial data name and last data name based on the sorting.

[0034] Further, the stage determination unit is used to transmit the basic time period, the binomial time period, the final time period, the basic data name, the binomial data name and the final data name to the acquisition control unit.

[0035] Furthermore, the specific method of determining the basic data name, the binomial data name and the last data name according to the sorting is as follows:

[0036] The top thirty percent are marked as basic data names, thirty to seventy percent are marked as binary data names, and the rest are marked as final data names.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention determines through a stage analysis unit that the same type of equipment will be in any of the basic period, the second period and the final period under different service durations, and then analyzes the content of the enabled monitoring object according to the different periods;

[0039] The monitoring object analyzes several maintenance data, observes the number of occurrences of each monitoring data under abnormal conditions, determines which monitoring objects appear frequently based on the number of occurrences, and thus determines to monitor different data at different stages; thereby, the present application can observe a small amount of data under different circumstances to determine whether an abnormality occurs. The present application is simple and effective, does not need to monitor all data, and can save computing power in some cases. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0041] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1 ,This application provides a hydropower unit health status evaluation system based on multi-source information fusion;

[0043] As the first embodiment of the present application, it specifically includes:

[0044] Data collection unit, stage determination unit, acquisition control unit, monitoring unit, multi-source analysis unit and display unit;

[0045] Among them, the data collection unit is used to collect data on the same hydropower units that have been scrapped in the past six months, and obtain the cycle information of the hydropower units. The cycle information specifically includes: maintenance times, the corresponding maintenance time for each time, and the total service time. The maintenance times here refer to the number of times the equipment of the corresponding hydropower unit is repaired before being scrapped, and the total service time refers to the total use time of the corresponding equipment. It is based on observing the total service time of several identical hydropower units after being scrapped and then taking the average value. The maintenance times are also determined by the average value of the equipment of several hydropower units;

[0046] The data collection unit is used to transmit the cycle information to the stage determination unit, and the stage determination unit is used to perform stage summary analysis based on the cycle information. The specific method of stage summary analysis is as follows:

[0047] Select any equipment of the same type as any equipment in the hydropower unit that has been scrapped in the past six months and mark it as the target object;

[0048] Get the total service duration of the target object;

[0049] Set a basic time period and mark it as T1. T1 is a preset value. The value of T1 generally does not exceed one third of the total service time.

[0050] Then, the number of times the corresponding target object is repaired in the basic period is obtained, which is marked as the maintenance segment; and the total maintenance time is marked as the maintenance segment time, and the maintenance segment time is divided by the maintenance segment number to obtain a value marked as the maintenance segment average time; the maintenance segment number and maintenance segment time are both determined by the average value of the values ​​of several scrapped equipment in the corresponding stage;

[0051] Then, the second third time period after the basic time period is obtained, that is, the time period corresponding to one third of the total service time after the basic time period, and it is marked as the binomial time period, and the rest is marked as the final time period; the maintenance time and maintenance times of the binomial time period and the final time period are obtained simultaneously;

[0052] Here, the maintenance time and maintenance times of the basic period, the two-term period and the final period are calculated by the average values ​​of several target objects of the same type in the corresponding different periods. Here, if a specific example is given, it means that one target object is selected, and then the maintenance times and maintenance time of several target objects of the same type in the basic period are obtained, and then the average values ​​of several maintenance times are calculated to represent the maintenance times of the target object;

[0053] The dimension segment average time of the second period and the final period is obtained, and marked as dimension segment average time 2 and dimension segment average time 3 respectively;

[0054] Then, the difference between the mean time of the second dimension segment and the mean time of the dimension segment is obtained, which is marked as the segment difference value. The basic period and the binomial period are adjusted according to the segment difference value. The specific method is as follows:

[0055] When X1≤segment difference≤X2, no processing is done at this time. X1 and X2 are both preset values, indicating that the division of the basic period and the binomial period is reasonable. If the segment difference is less than X1, the basic period will be automatically reduced by the set duration T2 in sequence. The reduction method is to push the end time point of the basic period forward by T2, and push the start time point of the binomial period forward by T2. The basic period and the binomial period are updated, and the segment difference is recalculated after the update. If the segment difference is still less than X1, the T2 duration will continue to be reduced, and the basic period and the binomial period will be updated until the segment difference satisfies X1≤segment difference≤X2;

[0056] If the segment difference is greater than X2, the end time point of the binomial period will be pushed forward by T2, and then the segment difference between the updated binomial period and the basic period will be recalculated. If it is still greater than X2, it will continue to be pushed forward by T2 until the segment difference satisfies X1≤segment difference≤X2;

[0057] Update the binomial period. When the basic period and binomial period are updated, the remaining period is marked as the last period.

[0058] Get the basic period, binomial period and final period;

[0059] The data collection unit is also used to collect the maintenance information of each corresponding target object reported for reimbursement in the past six months. The maintenance information includes the abnormal data name and abnormal duration. The abnormal data name is the name of the element that deviates when the corresponding abnormality occurs, such as stator temperature, rotor temperature, etc. The abnormal duration is the time required for maintenance and recovery after the corresponding abnormality occurs, and the abnormal data name and abnormal duration are transmitted to the stage determination unit. The stage determination unit is used to analyze the abnormal data name and abnormal duration for abnormal labels. The specific method of abnormal label analysis is as follows:

[0060] Select any maintenance information, assign the abnormal duration to the abnormal data name, and re-mark it as an abnormal value. Add up the abnormal values ​​of all the same abnormal data names, and mark the obtained value as the abnormal total value, so as to obtain the abnormal total value of each abnormal data name;

[0061] Get the number of times the abnormal data name appears, mark it as abnormal times, and use the formula to calculate the abnormal recognition value, abnormal recognition value = 0.62 × abnormal times + 0.38 × total abnormal value;

[0062] Sort the abnormal data names according to the total abnormal value, mark the top 30% as basic data names, mark 30% to 70% as binary data names, and the rest as final data names;

[0063] The stage determination unit is used to transmit the basic time period, the binomial time period, the final time period, the basic data name, the binomial data name and the final data name to the acquisition control unit;

[0064] The acquisition control unit is used to synchronize the monitored real-time object, and determine the real-time period it is in according to its working time. When the real-time period is in the basic period, the monitoring unit will be driven to obtain the real-time value of the data corresponding to the basic data name, and transmit the real-time data to the multi-source analysis unit to analyze whether the real-time object is abnormal, and display it through the display unit when an abnormality occurs. The multi-source analysis unit here uses the existing technology to monitor and analyze several different real-time data, which will not be described in detail here;

[0065] If the real-time period is in the binomial period, the monitoring unit will be driven to obtain the real-time values ​​of the data corresponding to the basic data name and the binomial data name, and the real-time data will be transmitted to the multi-source analysis unit to analyze whether the real-time object is abnormal, and display it through the display unit when an abnormality occurs. The multi-source analysis unit here uses the existing technology to monitor and analyze several different real-time data, which will not be described in detail here;

[0066] If the real-time time period is in the last time period, the monitoring unit will be driven to obtain the real-time values ​​of all data corresponding to the basic data name, the second data name and the last data name, and the real-time data will be transmitted to the multi-source analysis unit to analyze whether there is an abnormality in the real-time object, and display it through the display unit when an abnormality occurs. The multi-source analysis unit here uses existing technology to monitor and analyze several different real-time data, which will not be elaborated here.

[0067] Of course, as another embodiment of the present application, the present application is carried out on the basis of the above-mentioned embodiment. In this embodiment, the following method can be used to calculate and evaluate the efficiency of the unit, specifically by means of the value of the comprehensive evaluation index F, and F is calculated by the following method:

[0068] ,

[0069] η is the unit efficiency, x and y are the unit active power and working water head respectively;

[0070] H N is the working head, in m;

[0071] H ave is the weighted average working head (design value, a constant), unit: m.

[0072] ; At the same time, it is limited here ;

[0073] N g is the active power of the unit;

[0074] N e is the rated power of the unit;

[0075] w1 is the efficiency weight value;

[0076] w2 is the weight value of water consumption rate;

[0077] c is the weight coefficient, a user-configured value, a constant, and generally takes a value of 5;

[0078] k is the output margin coefficient, a configuration value, which is a constant and is generally taken as 1.02;

[0079] When the comprehensive evaluation index F is lower than the set threshold, it indicates an abnormality.

[0080] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The hydropower unit health status evaluation system based on multi-source information fusion is characterized by: include: The acquisition control unit is used to synchronize the monitored real-time object, determine its real-time period according to its working time, and perform status evaluation operations according to the real-time period. The specific operation method is as follows: When the real-time period is in the basic period, the monitoring unit will be driven to obtain the real-time value of the data corresponding to the basic data name, and transmit the real-time data to the multi-source analysis unit to analyze whether the real-time object is abnormal; If the real-time period is in the binomial period, the monitoring unit will be driven to obtain the real-time values ​​of the data corresponding to the basic data name and the binomial data name, and the real-time data will be transmitted to the multi-source analysis unit to analyze whether the real-time object is abnormal; If the real-time period is in the last period, the monitoring unit will be driven to obtain the real-time values ​​of all data corresponding to the basic data name, the second data name and the last data name, and the real-time data will be transmitted to the multi-source analysis unit to analyze whether the real-time object is abnormal; When the multi-source analysis unit finds an abnormality, it will transmit an abnormal signal to the display unit, and the display unit will automatically display the abnormal signal when it receives it; The basic period, the binomial period and the final period are determined by the stage determination unit through stage summary analysis. The specific method of stage analysis is as follows: Select any equipment of the same type as any equipment in the hydropower unit that has been scrapped in the past six months and mark it as the target object; The total service time and maintenance times of the target object are obtained. Here, the average total service time of all scrapped equipment is used to determine the maintenance times. Similarly, the maintenance times are also determined according to the average maintenance times of several scrapped hydropower units. First, the total service time is preliminarily divided into three stages, namely the basic period, the two-term period and the final period. Then, the number of maintenance times and the maintenance time of the corresponding equipment in each stage are calculated. The maintenance time is divided by the number of maintenance times to determine the average maintenance time. Then, according to the difference between the average maintenance times of two adjacent maintenance segments, that is, the segment difference, and the relationship between the segment difference and the set values ​​X1 and X2, the duration of the basic period, the two-term period and the final period is adjusted. After adjustment, determine the basic period, binomial period and final period.

2. The hydropower unit health status evaluation system based on multi-source information fusion according to claim 1 is characterized in that: The specific method for determining the basic period, the binomial period and the final period is as follows: Set a basic period and mark it as T1, where T1 is the preset value; Then, the number of times the corresponding target object is repaired in the basic period is obtained, which is marked as the maintenance segment number; and the total maintenance time is marked as the maintenance segment time, and the maintenance segment time is divided by the maintenance segment number to obtain a value marked as the average maintenance segment time; Then, the second one-third time period after the basic time period is obtained, that is, the time period corresponding to one-third of the total service time after the basic time period, and it is marked as the binomial time period, and the rest is marked as the final time period; The maintenance time and maintenance number of the second period and the last period are obtained synchronously; The maintenance time and maintenance times of the basic period, the two-term period and the final period are calculated by the average values ​​of several target objects of the same type in the corresponding different periods. Specifically, one target object is selected, and then the maintenance times and maintenance time of several target objects of the same type in the basic period are obtained, and then the average values ​​of several maintenance times are calculated to represent the maintenance times of the target object. The dimension segment average time of the second period and the final period is obtained, and marked as dimension segment average time 2 and dimension segment average time 3 respectively; Then, the difference between the mean time of the second dimension segment and the mean time of the dimension segment is obtained, marked as the segment difference value, and the basic period and the binomial period are adjusted according to the segment difference value; Update the binomial period. When the basic period and binomial period are updated, the remaining period is marked as the last period. Get the basic period, binomial period and final period.

3. The hydropower unit health status evaluation system based on multi-source information fusion according to claim 2 is characterized in that: The specific method of adjusting the basic period and the binomial period according to the segment difference value is: When X1≤segment difference≤X2, no processing is done at this time. X1 and X2 are both preset values, indicating that the division of the basic period and the binomial period is reasonable. If the segment difference is less than X1, the basic period will be automatically reduced by the set duration T2 in sequence. The reduction method is to push the end time point of the basic period forward by T2, and push the start time point of the binomial period forward by T2. The basic period and the binomial period are updated, and the segment difference is recalculated after the update. If the segment difference is still less than X1, the T2 duration will continue to be reduced, and the basic period and the binomial period will be updated until the segment difference satisfies X1≤segment difference≤X2; If the segment difference is greater than X2, the end time point of the binomial period will be pushed forward by T2, and then the segment difference between the updated binomial period and the basic period will be recalculated. If it is still greater than X2, it will continue to be pushed forward by T2 until the segment difference satisfies X1≤segment difference≤X2.

4. The hydropower unit health status evaluation system based on multi-source information fusion according to claim 1 is characterized in that: It also includes a data collection unit for collecting data on the same hydropower units that have been scrapped in the past six months, and obtaining the cycle information of the hydropower units. The cycle information specifically includes: maintenance times, the duration of each maintenance, and the total service duration. The maintenance times refer to the number of times the equipment of the corresponding hydropower unit was repaired before being scrapped, and the total service duration refers to the total usage time of the corresponding equipment. It is based on observing the total service duration of several identical hydropower units after being scrapped and then taking the average value. The maintenance times are also determined by the average of the equipment of several hydropower units.

5. The hydropower unit health status evaluation system based on multi-source information fusion according to claim 1 is characterized in that: The data collection unit is also used to collect each maintenance information of the corresponding target object reimbursed in the past six months. The maintenance information includes the abnormal data name and abnormal duration. The abnormal data name is the name of the element that deviates when the corresponding abnormality occurs, and the abnormal duration is the duration required for maintenance and recovery after the corresponding abnormality occurs. The abnormal data name and abnormal duration are transmitted to the stage determination unit. The stage determination unit is used to analyze the abnormal data name and abnormal duration for abnormal labels. The specific method of abnormal label analysis is as follows: Select any maintenance information, assign the abnormal duration to the abnormal data name, and re-mark it as an abnormal value. Add up the abnormal values ​​of all the same abnormal data names, and mark the obtained value as the abnormal total value, so as to obtain the abnormal total value of each abnormal data name; Get the number of times the abnormal data name appears, mark it as abnormal times, and use the formula to calculate the abnormal recognition value, abnormal recognition value = 0.62 × abnormal times + 0.38 × total abnormal value; Sort the abnormal data names according to the total abnormal value, and determine the basic data name, binomial data name and last data name based on the sorting.

6. The hydropower unit health status evaluation system based on multi-source information fusion according to claim 1 is characterized in that: The phase determination unit is used to transmit the basic time period, the binomial time period, the final time period, the basic data name, the binomial data name and the final data name to the acquisition control unit.

7. The hydropower unit health status evaluation system based on multi-source information fusion according to claim 5 is characterized in that: The specific method for determining the basic data name, the binomial data name and the last data name according to the sorting is: The top thirty percent are marked as basic data names, thirty to seventy percent are marked as binary data names, and the rest are marked as final data names.

Images