CN117870749A - Sensor performance evaluation method, device, electronic equipment and medium - Google Patents
Sensor performance evaluation method, device, electronic equipment and medium Download PDFInfo
- Publication number
- CN117870749A CN117870749A CN202410057671.5A CN202410057671A CN117870749A CN 117870749 A CN117870749 A CN 117870749A CN 202410057671 A CN202410057671 A CN 202410057671A CN 117870749 A CN117870749 A CN 117870749A
- Authority
- CN
- China
- Prior art keywords
- time window
- deviation value
- sensor
- sampling
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D18/00—Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/02—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance
- G01N27/04—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance
Landscapes
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses a sensor performance evaluation method, a device, electronic equipment and a medium, wherein the method comprises the following steps: acquiring first data actually acquired by a sensor at each sampling moment and second data acquired in a simulation manner; configuring multiple types of time windows for each sampling time, and respectively determining a first characteristic and a second characteristic corresponding to each time window configured for each sampling time according to the first data and the second data in each time window of each sampling time; determining a deviation value of each time window in the sensor sampling process according to the first characteristic and the second characteristic corresponding to each time window configured at each sampling moment; and determining the performance score of the sensor according to the deviation value corresponding to each time window. By adopting the method, the performance of the evaluation sensor can be monitored comprehensively and accurately from the perspective of full sample evaluation through full-quantity and real sensor sampling data.
Description
Technical Field
The present invention relates to the field of sensor technologies, and in particular, to a sensor performance evaluation method, a device, an electronic apparatus, and a medium.
Background
Along with the rapid development of urban lifeline engineering and industrial Internet ecological projects, various sensors are distributed in all corners to operate, various parameters are collected in real time through a large number of sensors and uploaded to a cloud, and data calculation of various businesses and various models is carried out, so that various monitoring operation services are provided for the urban lifeline engineering. In this business scenario, all ecological applications running in urban safety monitoring must be effectively run based on sensor data, so that it is necessary to ensure that the data collected by the sensors are true and reliable. In the related art, quality inspection of a sensor mainly depends on manual sampling and detection, and has many problems, such as too small sample number, complicated laboratory inspection process, easy misjudgment of results and the like, and the sensor performance cannot be accurately and comprehensively monitored and evaluated.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a sensor performance evaluation method, apparatus, electronic device, and medium capable of comprehensively and accurately evaluating sensor performance.
A method of sensor performance evaluation, comprising: acquiring first data actually acquired by a sensor at each sampling moment and second data acquired in a simulation manner;
Configuring multiple types of time windows for each sampling time, and respectively determining a first characteristic and a second characteristic corresponding to each time window configured for each sampling time according to the first data and the second data in each time window of each sampling time; wherein, the corresponding set width of each type of time window is different;
determining a deviation value of each time window in the sensor sampling process according to the first characteristic and the second characteristic corresponding to each time window configured at each sampling moment;
in the above solution, when determining the deviation value of each time window in the sensor sampling process according to the first feature and the second feature corresponding to each time window configured at each sampling time, the method includes:
sequentially determining the corresponding deviation value of each time window according to the set sequence; the set order is determined based on the width of each type of time window from the large to the small order.
In the above solution, when determining the deviation value of each time window in the sensor sampling process according to the first feature and the second feature corresponding to each time window configured at each sampling time, the method includes:
Comparing the deviation value corresponding to the determined time window with a preset deviation value corresponding to the type of the time window; wherein the preset deviation value is preset for each type of time window;
retaining the deviation value if the deviation value is determined to be less than or equal to the preset deviation value;
and under the condition that the deviation value is larger than the preset deviation value, updating the deviation value corresponding to the time window into a set value.
In the above scheme, the determining the deviation value corresponding to each time window sequentially according to the set order includes:
comparing the deviation value corresponding to the currently determined time window with a preset deviation value corresponding to the type of the time window; wherein the preset deviation value is preset for each type of time window;
determining a deviation value corresponding to a next type of time window according to the setting sequence under the condition that the deviation value is smaller than or equal to the preset deviation value;
and stopping calculating the deviation values corresponding to other types of time windows positioned behind the currently determined time window under the condition that the deviation value is determined to be larger than the preset deviation value.
In the above solution, in the case that it is determined that the deviation value is greater than the preset deviation value, the method further includes:
and determining the deviation value corresponding to the currently determined time window and other types of time windows positioned behind the currently determined time window as a set value.
In the above solution, the determining, according to the first feature and the second feature corresponding to each time window configured at each sampling time, a deviation value of each time window in the sensor sampling process includes:
according to the first features and the second features corresponding to each time window configured at each sampling moment, determining vector distances corresponding to each time window configured at each sampling moment;
and summing the vector distances corresponding to the time windows of the same type in all the sampling moments to obtain the deviation value corresponding to each time window.
In the above solution, the determining the performance of the sensor according to the deviation value corresponding to each time window includes:
adding the deviation values corresponding to each type of time window to obtain an evaluation score of the sensor performance; wherein the evaluation score is inversely related to the performance of the sensor.
A sensor performance evaluation device, comprising:
the acquisition module is used for acquiring the first data actually acquired by the sensor at each sampling moment and the second data acquired in a simulation way;
the first determining module is used for configuring multiple types of time windows for each sampling moment, and respectively determining a first characteristic and a second characteristic corresponding to each time window configured for each sampling moment according to the first data and the second data in each time window of each sampling moment; wherein, the corresponding set width of each type of time window is different;
the second determining module is used for determining the deviation value of each time window in the sensor sampling process according to the first characteristic and the second characteristic corresponding to each time window configured at each sampling moment;
and the evaluation module is used for determining the performance score of the sensor according to the deviation value corresponding to each time window.
An electronic device comprising a memory storing a computer program and a processor implementing the steps of the above-described sensor performance evaluation method when the processor executes the computer program.
A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the above-described sensor performance evaluation method.
According to the sensor performance evaluation method, the sensor performance evaluation device, the electronic equipment and the sensor medium, the first characteristics and the second characteristics in different types of windows are extracted through the actual acquired data and the simulated acquired data of different types of time windows at different sampling moments, the deviation values of the first characteristics and the second characteristics in each time window are determined by utilizing the first characteristics and the second characteristics corresponding to the time windows, so that the performance score of the sensor is obtained, and the performance of the sensor is comprehensively and accurately monitored and evaluated through the full-quantity and real sensor sampling data from the perspective of full-sample evaluation.
Drawings
FIG. 1 is a flow diagram of a method of sensor performance evaluation in one embodiment;
FIG. 2 is a flowchart illustrating determining a deviation value corresponding to a time window according to an embodiment;
FIG. 3 is a flowchart illustrating a method for determining a deviation value corresponding to a time window according to another embodiment;
FIG. 4 is a flowchart illustrating a method for determining a deviation value corresponding to a time window according to another embodiment;
FIG. 5 is a block diagram of a sensor performance evaluation apparatus in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
Implementation details of the technical solutions of the embodiments of the present application are described in detail below.
In one embodiment, as shown in FIG. 1, a sensor performance evaluation method is provided, which may include the steps of:
step S101, acquiring first data actually acquired by the sensor at each sampling time and second data acquired by simulation.
In this embodiment, there are two parts of collected data, the first part of data is the first data actually collected by the sensor at each sampling time in the actual working process, and the second part of data is the second data collected by the analog sensor at each sampling time under the normal working condition, that is, the second data is generated through simulation, and is the theoretical value of the signal collected by the sensor.
The second data acquired by the acquisition sensor in a simulation will be described below by taking a gas methane detection sensor as an example.
When the analog sensor collects data, a working principle report of the gas methane detection sensor needs to be acquired firstly, wherein the working principle of the gas methane detection sensor is generally based on the semiconductor gas sensor, semiconductor materials in the gas methane detection sensor can chemically react with methane gas to cause the change of the resistance of the semiconductor, the change of the resistance can be converted into a voltage signal to be output, and the higher the concentration of the methane gas is, the larger the resistance change is, so that the output voltage signal is in direct proportion to the concentration of the methane. Assuming that the sensitivity of the methane gas detection sensor is S and the methane gas concentration is C, the voltage signal output by the sensor may be expressed as v=s×c, and this formula may be used to simulate the theoretical value of the pilot acquisition signal of the methane gas detection sensor under the theoretical normal working condition.
In the process of data acquisition of the analog sensor, different working environments are also required to be designed for the sensor, so that data acquisition of the gas methane detection sensor in the normal working state under different working environments can be simulated. The following is a simulation of data acquisition corresponding to gas methane detection sensors in different working environments:
(1) The gas leakage rises slowly: under the working environment, the methane gas concentration can slowly rise, the response time of the gas methane detection sensor is long, the response curve of the gas methane detection sensor can be simulated, the methane concentration is assumed to be increased by a fixed amount at intervals, and then the corresponding voltage signal value is calculated according to the response characteristic of the gas methane detection sensor.
(2) The gas leakage rises rapidly: under the working environment, the concentration of methane gas can rise rapidly, the response time of the gas methane detection sensor is short, the concentration of methane can be simulated to rise rapidly to a high value, and then the corresponding voltage signal value is calculated according to the response characteristic of the gas methane detection sensor.
(3) Signal value is unstable: in such an operating environment, environmental factors such as temperature, humidity, etc. may cause the signal value output from the gas methane detection sensor to fluctuate. The corresponding signal value fluctuation can be calculated by simulating the change of environmental factors, such as the periodic change of temperature.
In practical applications, the data collection of the analog sensor can be performed by computer simulation software or a mathematical simulation tool.
It should be noted that, the acquired first data and the second data are one time sequence signal value, where the acquired first data may be expressed as: t (T) 1 ：M 1 ，T 2 ：M 2 ，…，T m ：M m Wherein T is 1 ，T 2 ，…，T m Represents the sampling time, M 1 ，M 2 ，…，M m Representing first data, i.e. sensor at T 1 M is actually acquired at the moment 1 。
The acquired second data may be expressed as: t (T) 1 ：m 1 ，T 2 ：m 2 ，…，T m ：m m Wherein T is 1 ，T 2 ，…，T m Represents the sampling time, m 1 ，m 2 ，…，m m Representing second data, i.e. during simulation of the sampling of the sensor, the sensor is at T 1 M is acquired at the moment 1 。
Step S102, configuring multiple types of time windows for each sampling moment, and respectively determining a first feature and a second feature corresponding to each time window configured for each sampling moment according to first data and second data in each time window of each sampling moment.
Here, different types of time windows are set, the types of which are distinguished by the width of the time window, and the types of the time windows are classified into: 1h,2h,4h,8h,12h,24h,2day,4day,1week and 2week, wherein a time window of 1h is used to frame data within 1 hour, a time window of 2day is used to frame data within 2 days, and a time window of 1week is used to frame data within 1 week. In practical applications, the type of the time window may be adjusted according to practical needs, and in this embodiment, the above-listed 10 types of time windows are taken as an example for illustration.
In this embodiment, a plurality of types of time windows are configured for each sampling instant, where the types of time windows include the 10 types of time windows listed above.
After configuring a time window for each sampling instant, there will be corresponding first and second data falling within the time window, for example, assuming T 1 One type of time window is configured as a time window of 1h, then the time window is set to T 1 As the starting time, the current T is obtained 1 First data and second data of a previous 1h length.
Next, the first data and the second data that fall within each time window configured for each sampling instant need to be processed. In practical applications, the processing flows for the first data and the second data within each time window are identical, except that the amount of data processed is different. In order to more clearly describe the processing flow of the first data and the second data in the time window, a time window corresponding to one of the sampling moments is described as an example.
And processing the first data and the second data falling into the time window by utilizing the characteristic function, wherein the first characteristic of the first data in the time window can be calculated through processing the first data, and the second characteristic of the second data in the time window can be calculated through processing the second data.
In practical applications, a plurality of different feature functions may be used in the process of extracting the data features, so that different feature extraction may be performed on the first data in the time window. In practical applications, the feature function may include different statistical functions such as mean, variance, maximum, minimum, maximum slope, minimum slope, etc., for extracting distribution features and statistical features of data.
The following is a first characteristic obtained by a time window of 1h at different sampling moments
T 1 ：f1_t 1 ，f2_t 1 ，…fn_t 1
T 2 ：f1_t 2 ，f2_t 2 ，…fn_t 2
……
T m ：f1_t m ，f2_t m ，…fn_t m
Wherein, f1, f2, … …, fn refer to the calculated values of each feature function, the feature functions are N in number, and for example, f1 may correspond to the mean value in the time window, and f2 may correspond to the variance in the time window; subscript t 1 ，t 2 ，……t m Corresponding to the sampling instant, i.e. at T 1 The first characteristic obtained in the time window of 1h corresponding to the sampling moment is T 1 ：f1_t 1 ，f2_t 1 ，…fn_t 1 At T 2 The first characteristic obtained in the time window of 1h corresponding to the sampling moment is T 2 ：f1_t m ，f2_t m ，…fn_t m 。
The following is the second characteristic obtained by the time window of 1h at different sampling moments:
T 1 ：f1_t 1 _M，f2_t 1 _M，…fn_t 1 _M
T 2 ：f1_t 2 _M，f2_t 2 _M，…fn_t 2 _M
……
T m ：f1_t m _M，f2_t m _M，…fn_t m _M
wherein, f1, f2, … …, fn refer to the calculated values of each feature function, the feature functions are N in number, and for example, f1 may correspond to the mean value in the time window, and f2 may correspond to the variance in the time window; subscript t 1 ，t 2 ，……t m Corresponding to the sampling instant, i.e. at T 1 In a time window of 1h corresponding to the sampling moment, the obtained second characteristic is f1_t 1 _M，f2_t 1 _M，…fn_t 1 M, at T 2 The second characteristic obtained in the time window of 1h corresponding to the sampling moment is T 2 ：f1_t 2 _M，f2_t 2 _M，…fn_t 2 _M。
And carrying out corresponding processing on each time window configured at each sampling moment to obtain a first characteristic and a second characteristic corresponding to each time window configured at each sampling moment, wherein the first characteristic refers to distribution and statistical characteristics of first data in the time window, and the second characteristic refers to distribution and statistical characteristics of second data in the time window.
In practical application, the feature function may be a User-defined function (UDF), and the User can use the UDF function to extract features of data in the process of data processing by writing a function for extracting features of data and registering the written function as the UDF. The UDF may include, in addition to statistical functions, physical model functions, where the physical model functions may involve functions related to various physical laws, mathematical models, and the like, such as a linear model, a nonlinear model, and a differential equation.
In one implementation, logic for defining data stream processing within a time window in a Flink computing frame may be utilized, including setting different types of time windows in the Flink computing frame, such that the Flink computing frame obtains data of window lengths before different sampling moments using the different types of time windows, and setting UDF functions in the Flink computing frame, wherein during processing of the data stream, the first data and the second data are feature extracted using the UDF functions.
In this embodiment, the multiple types of time windows are used to observe the change situations of the first data and the second data in different time units, so that the first data and the second data collected by the sensor can be more comprehensively analyzed.
In practice, after the first and second features are determined, the first and second features can be stored in a time-series database, wherein the time-series database is a database system dedicated to storing and managing time-series data. It is designed to process large-scale time-series data, such as sensor data, and the time-series database has efficient data storage and query capability, can rapidly process a large amount of time-series data, and provides flexible data analysis and query functions.
In storing the first and second features in the time sequence database, a storage principle of "one-object-one-table" is required to be followed, wherein "one-object-one-table" refers to that a separate data table is created for each sensor to store the data of the sensor, and the storage principle can enable the data of each sensor to have own independent storage space in the time sequence database, so that management and query are convenient.
After the first features and the second features are stored in the time sequence database, when the first features and the second features corresponding to different time windows at different sampling moments are used for subsequent data processing, corresponding data can be rapidly extracted from the time sequence database.
Step S103, determining the deviation value of each time window in the sensor sampling process according to the first characteristic and the second characteristic corresponding to each time window configured at each sampling time.
Here, comparing the first feature and the second feature in each time window configured according to each sampling time, and finally determining a deviation value of each time window, specifically, taking a time window of 1h as an example, configuring a time window of 1h at each sampling time, so as to process by using the first feature and the second feature in the time window of 1h corresponding to each sampling time, and determining a data deviation condition of the sensor in the whole sampling process in a unit time of 1h, wherein the deviation value is used for evaluating a similarity degree of the first data and the second data of the sensor in the sampling process, and the similarity degree is obtained by comparing the corresponding first feature and the second feature.
Based on the above, the deviation values of the above-listed 10 time windows in the sensor sampling process can be obtained, so that the fluctuation or instability of the sensor in the data acquisition process can be monitored through the deviation values corresponding to different time windows.
In one embodiment, in determining the deviation value of each time window, the setting sequence is determined according to the order of the widths corresponding to each type of time window from large to small, and, for example, assuming that the types of the time windows are 10 types listed above, the corresponding sequence is 2week, 1week, 4day, 2day, 24h, 12h, 8h, 4h, 2h, 1h, that is, the deviation value of the 2week time window configured at each sampling time is determined first, and the deviation value of the 1h time window configured at each sampling time is determined finally.
In this embodiment, the deviation values of the time windows are determined from the sequence of the large to small time windows, so that the whole trend can be started, and then the time windows gradually go deep into smaller time scales, which is helpful for comprehensively understanding the change rule of the sensor in the data acquisition process.
In one embodiment, as shown in fig. 2, fig. 2 shows a flow chart for determining the deviation value corresponding to the time window.
Step S201, comparing the deviation value corresponding to the determined time window with a preset deviation value corresponding to the type of the time window.
In step S202, in the case where it is determined that the deviation value is less than or equal to the preset deviation value, the deviation value is retained.
In step S203, when the deviation value is determined to be greater than the preset deviation value, the deviation value corresponding to the time window is updated to the set value.
After determining the deviation value corresponding to the time window, comparing the deviation value corresponding to the time window with a preset deviation value preset for the time window.
It should be noted that, the preset deviation value is preset for different types of time windows, for example, for a time window of 1h, the corresponding preset deviation value may be set to t_1h, for a time window of 2week, the corresponding preset deviation value may be set to t_2week, so that the preset deviation value corresponding to the time window is substantially determined according to the type of the time window, and if the time window is a time window of 2week, the determined deviation value d_2week corresponding to the time window is compared with the preset deviation value t_2week.
If the deviation value D_2week corresponding to the time window is less than or equal to the preset deviation value T_2week, the deviation value D_2week is considered to be within a reasonable range, and the calculated deviation value D_2week is reserved.
If the deviation value d_2week corresponding to the time window is greater than the preset deviation value t_2week, the deviation value d_2week is considered to be a bad value, that is, the deviation value d_2week is beyond a reasonable range, the deviation value d_2week corresponding to the time window is updated to be a set value, wherein the set value is a larger value, which can be regarded as an infinite value, and the deviation value beyond the reasonable range can be obviously distinguished from the deviation value within the reasonable range, so that the time window with larger deviation at the position can be obviously marked, and the time window can be more easily identified and processed in the subsequent process of calculating the performance score of the sensor.
As shown in table 1, table 1 shows the deviation values of each type of time window of the sensor.
| Time window | Deviation value |
| 2week | G_2week |
| 1week | G_1week |
| 4day | G_4day |
| 2day | Setting value |
| 24hour | Setting value |
| 12hour | Setting value |
| 8hour | Setting value |
| 4hour | Setting value |
| 2hour | Setting value |
| 1hour | Setting value |
From table 1, it can be determined that, starting from the 2day time window, the corresponding deviation value is out of the reasonable range, based on which the deviation value of the 2day time window and the following time windows is set as the set value, while the deviation value of the time window preceding the 2day time window is still the calculated deviation value.
In this embodiment, the comparison between the determined offset value corresponding to the time window and the preset offset value may be performed after all the offset values corresponding to the time windows are determined, or may be performed after the offset value corresponding to one time window is determined, and the offset value corresponding to the next time window is determined after the comparison is completed.
In one embodiment, as shown in fig. 3, fig. 3 shows a flow chart for determining the deviation value corresponding to the time window.
Step S301, comparing the deviation value corresponding to the currently determined time window with a preset deviation value corresponding to the type of the time window.
In step S302, in the case where it is determined that the deviation value is less than or equal to the preset deviation value, the deviation value corresponding to the next type of time window is determined in the set order.
In step S303, in the case where the deviation value is determined to be greater than the preset deviation value, calculation of the deviation value corresponding to the other type of time window located after the currently determined time window is stopped.
In the process of determining the offset values corresponding to the time windows according to the set sequence, after determining the offset value corresponding to the current time window, the offset value corresponding to the current time window is compared with the corresponding preset offset value, and whether the offset value corresponding to the subsequent time window needs to be determined can be determined according to the comparison result. The following is a description by way of specific examples.
According to the setting sequence, firstly determining a deviation value corresponding to a time window of 2week, and after determining a deviation value D_2week corresponding to the time window of 2week, comparing the deviation value D_2week with a preset deviation value T_2week corresponding to the time window of 2week to determine whether the deviation value D_2week is within a reasonable range.
Wherein the preset deviation value is preset for each type of time window, i.e. the preset deviation values of the same type of time window are the same.
Under the condition that the deviation value D_2week is less than or equal to the preset deviation value T_2week, namely the deviation value D_2week is in a reasonable range, and further, the preset deviation value corresponding to the subsequent time window is determined according to the set sequence.
In case it is determined that the deviation value d_2week is larger than the preset deviation value t_2week, that is, the deviation value d_2week is out of a reasonable range. It can be understood that determining the deviation value corresponding to the time window according to the set order determines the deviation degree between the first data and the second data from the order of the large time dimension to the small time dimension, so that, when the deviation value corresponding to the larger time window is larger than the preset deviation value, the deviation value corresponding to the smaller time window obtained in the subsequent calculation process is also beyond a reasonable range, and based on this, it is not significant to further calculate the deviation value of the subsequent time window, so that in this case, calculation of the deviation value of other types of time windows after the currently determined time window is suspended, and thus the subsequent data analysis and processing steps can be simplified, and a great amount of time and calculation resources can be saved.
In one embodiment, in the case that the deviation value corresponding to the currently determined time window is greater than the preset deviation value, the determination of the deviation value corresponding to the subsequent time window is suspended, based on which, in this embodiment, the deviation values corresponding to the currently determined time window and other types of time windows located after the currently determined time window are determined as set values, wherein the set values are almost infinite, and, for example, assuming that the width of the currently determined time window is 4h, the deviation values corresponding to the 4h time window, the 2h time window, and the 1h time window are all set as set values.
It will be appreciated that the width of the preceding time window in the set sequence will be greater than the width of the following time window, i.e. the offset value is determined from a larger time dimension before further determining the offset value from a smaller time dimension. Based on the above, the deviation value determined in the larger time dimension exceeds the reasonable range, and the deviation value determined in the microcosmic condition also exceeds the reasonable range, so that the deviation value corresponding to the currently determined time window and other types of time windows positioned behind the currently determined time window is set as a set value, the deviation value exceeding the reasonable range is obviously distinguished from the deviation value within the reasonable range by the set value, and the time window with the larger deviation at the position can be obviously marked, so that the deviation value can be more easily identified and processed in the process of calculating the performance score of the sensor.
In one embodiment, as shown in fig. 4, fig. 4 shows a flow chart for determining the deviation value corresponding to the time window.
Step S401, determining a vector distance corresponding to each time window configured at each sampling time according to the first feature and the second feature corresponding to each time window configured at each sampling time.
Step S402, summing the vector distances corresponding to the time windows of the same type in all sampling moments to obtain the deviation value corresponding to each time window.
Here, the vector pitch between the first feature and the second feature is evaluated based on the first feature and the second feature corresponding to each time window.
The following is an example of a time window of 2 week. For T 1 A time window of 2week configured at a moment, the corresponding first feature is expressed as: t (T) 1 ：f1_t 1 ，f2_t 1 ，…fh_t 1 The corresponding second feature is expressed as: t (T) 1 ：f1_t 1 _M，f2_t 1 _M，…fn_t 1 M, in one implementation, the vector spacing between the first feature and the second feature is the square of the euclidean distance, i.e., the spacing between the first feature and the second feature can be represented by calculating the square of the modulo length of their difference vectors, the corresponding operation expression being represented as:
by processing each time window configured for each sampling time according to the above-described operation formula, a vector distance corresponding to each time window configured for each sampling time can be determined.
In practical application, different types of time windows are configured for different sampling moments, so as to observe the difference between the first data and the second data in different unit time, and after the vector space corresponding to each time window configured for each sampling moment is determined, the overall change condition among feature vectors in the whole time sequence data needs to be comprehensively evaluated.
Based on the above, the vector distances corresponding to the time windows of the same type in all sampling moments are summed, so that the deviation value corresponding to each time window can be obtained, and the deviation value corresponding to the time window can reflect the overall difference between the feature vectors of the time windows in the time sequence data. Taking a time window of 2week as an example, the time window of 2week corresponds to the deviation valueThat is, the offset values of the 2week time windows corresponding to all sampling moments are added to obtain d_2week, and the offset value corresponding to each type of time window can be obtained finally.
Step S104, determining the performance scores of the sensors according to the deviation values corresponding to each time window.
The deviation value is used for measuring the deviation degree between the first data and the second data, it can be understood that the second data is a theoretical value obtained by simulating the data collected by the sensor, the second data is a reference standard for the performance of the sensor, and the expected value is calculated according to the design principle, physical characteristics or theoretical model of the sensor, that is, the actual data collected by the sensor needs to be close to the theoretical data obtained by simulation, so that the data collected by the sensor is more true and reliable under the condition that the first data is closer to the second data.
Based on the method, the deviation values in different time windows are comprehensively considered, the performance scores of the sensors can be determined by adopting a weighted average or other comprehensive methods, and in practical application, the performance scores can be used for quantifying the advantages and disadvantages of the sensor performance, so that the standardized evaluation of the sensor performance can be realized, and the influence of subjective factors is avoided. The performance scores herein are used primarily to evaluate the overall stability and accuracy of sensor data acquisition.
In one embodiment, an evaluation function of sensor performance is setWherein E is the performance score of the sensor, i is the type of time window, D -i The deviation values corresponding to the time windows of type i are indicated, i.e. the deviation values corresponding to the above listed 10 types of time windows are summed. It will be appreciated that the offset value is a measure of the degree of offset between the first data and the second data, and therefore, when the evaluation score E value is smaller, the closer the first data actually collected by the sensor is to the second data obtained by simulation, the better the sensor performance is.
It should be noted that, in the case where the deviation value corresponding to some time windows is a set value, the set value is a value that is almost infinite, in this case, the performance score calculated according to the evaluation function is also a value that is almost infinite, and in particular, in the case where the deviation value corresponding to the time window that is earlier in the set sequence is a set value, the performance score corresponding to the sensor is far greater than the performance score of the sensor for which the deviation value is not set as the set value, so that the quality of the sensor performance can be clearly distinguished.
It should be noted that the sensor performance evaluation method can be suitable for sensors of different manufacturers, different models and different batches, can accurately compare the quality of various types of sensors, and provides a solid theoretical support for the inferior ecological support of the whole sensor industry.
In the above embodiment, by acquiring the first data actually acquired by the sensor and the second data obtained by simulation, and comparing the difference conditions between the first data and the second data by using different time windows, the performance of the sensor can be evaluated by the deviation values corresponding to different time windows at different sampling moments, and the running state of the sensor can be comprehensively monitored and evaluated by the full-scale and real sensor sampling data.
In one embodiment, a sensor performance evaluation apparatus is provided, and referring to fig. 5, the sensor performance evaluation apparatus 500 may include: the system comprises an acquisition module 501, a first determination module 502, a second determination module 503 and an evaluation module 504.
The acquisition module 501 is configured to acquire first data actually acquired by the sensor at each sampling time and second data acquired by simulation; a first determining module 502, configured to configure multiple types of time windows for each sampling time, and determine, according to first data and second data in each time window of each sampling time, a first feature and a second feature corresponding to each time window configured for each sampling time; wherein, the corresponding set width of each type of time window is different; a second determining module 503, configured to determine a deviation value of each time window in the sensor sampling process according to the first feature and the second feature corresponding to each time window configured at each sampling time; an evaluation module 504 is configured to determine a performance score of the sensor according to the deviation value corresponding to each time window.
In one embodiment, the second determining module 503 is specifically configured to sequentially determine, according to a set order, a deviation value corresponding to each time window; the setting order is determined based on the width of each type of time window from the large to the small order.
In one embodiment, the second determining module 503 is specifically configured to compare the determined deviation value corresponding to the time window with a preset deviation value corresponding to the type of the time window; wherein the preset deviation value is preset for each type of time window;
under the condition that the deviation value is smaller than or equal to a preset deviation value, the deviation value is reserved;
and under the condition that the deviation value is larger than the preset deviation value, updating the deviation value corresponding to the time window into the set value.
In one embodiment, the second determining module 503 is specifically configured to compare the deviation value corresponding to the currently determined time window with a preset deviation value corresponding to the type of the time window; wherein the preset deviation value is preset for each type of time window;
under the condition that the deviation value is smaller than or equal to a preset deviation value, determining a deviation value corresponding to a next type of time window according to a set sequence;
And stopping calculating the deviation values corresponding to other types of time windows positioned behind the currently determined time window under the condition that the deviation value is determined to be larger than the preset deviation value.
In one embodiment, the second determining module 503 is further configured to determine, as the set value, a deviation value corresponding to the currently determined time window and other types of time windows located after the currently determined time window, in a case where the deviation value is determined to be greater than the preset deviation value.
In one embodiment, the second determining module 503 is specifically configured to determine, according to the first feature and the second feature corresponding to each time window configured at each sampling time, a vector distance corresponding to each time window configured at each sampling time;
and summing the vector distances corresponding to the time windows of the same type in all sampling moments to obtain the deviation value corresponding to each time window.
In one embodiment, the evaluation module 504 is specifically configured to add the deviation values corresponding to each type of time window to obtain an evaluation score of the sensor performance; wherein the evaluation score is inversely related to the performance of the sensor.
For specific limitations on the sensor performance evaluation device, reference may be made to the above limitations on the sensor performance evaluation method, and no further description is given here. The respective modules in the above-described sensor performance evaluation apparatus may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, an electronic device is provided that includes a memory storing a computer program and a processor that when executing the computer program implements a sensor performance assessment method.
In one embodiment, a computer storage medium having a computer program stored thereon, which when executed by a processor, implements a sensor performance evaluation method.
It should be noted that the logic and/or steps represented in the flowcharts or otherwise described herein, for example, may be considered as a ordered listing of executable instructions for implementing logical functions, and may be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.
Claims (10)
1. A method of evaluating sensor performance, comprising:
acquiring first data actually acquired by a sensor at each sampling moment and second data acquired in a simulation manner;
configuring multiple types of time windows for each sampling time, and respectively determining a first characteristic and a second characteristic corresponding to each time window configured for each sampling time according to the first data and the second data in each time window of each sampling time; wherein, the corresponding set width of each type of time window is different;
Determining a deviation value of each time window in the sensor sampling process according to the first characteristic and the second characteristic corresponding to each time window configured at each sampling moment;
and determining the performance score of the sensor according to the deviation value corresponding to each time window.
2. The sensor performance evaluation method according to claim 1, wherein when determining a deviation value of each time window during the sensor sampling from the first feature and the second feature corresponding to each time window configured for each sampling time, the method comprises:
sequentially determining the corresponding deviation value of each time window according to the set sequence; the set order is determined based on the width of each type of time window from the large to the small order.
3. The sensor performance evaluation method according to claim 1 or 2, wherein when determining a deviation value of each time window during the sensor sampling from the first feature and the second feature corresponding to each time window configured for each sampling time instant, the method comprises:
comparing the deviation value corresponding to the determined time window with a preset deviation value corresponding to the type of the time window; wherein the preset deviation value is preset for each type of time window;
Retaining the deviation value if the deviation value is determined to be less than or equal to the preset deviation value;
and under the condition that the deviation value is larger than the preset deviation value, updating the deviation value corresponding to the time window into a set value.
4. The sensor performance evaluation method according to claim 2, wherein the sequentially determining the deviation value corresponding to each time window in the set order comprises:
comparing the deviation value corresponding to the currently determined time window with a preset deviation value corresponding to the type of the time window; wherein the preset deviation value is preset for each type of time window;
determining a deviation value corresponding to a next type of time window according to the setting sequence under the condition that the deviation value is smaller than or equal to the preset deviation value;
and stopping calculating the deviation values corresponding to other types of time windows positioned behind the currently determined time window under the condition that the deviation value is determined to be larger than the preset deviation value.
5. The sensor performance evaluation method according to claim 4, wherein in the case where it is determined that the deviation value is greater than the preset deviation value, the method further comprises:
And determining the deviation value corresponding to the currently determined time window and other types of time windows positioned behind the currently determined time window as a set value.
6. The sensor performance evaluation method according to claim 1 or 2, wherein the determining a deviation value of each time window in the sensor sampling process from the first feature and the second feature corresponding to each time window configured for each sampling time instant includes:
according to the first features and the second features corresponding to each time window configured at each sampling moment, determining vector distances corresponding to each time window configured at each sampling moment;
and summing the vector distances corresponding to the time windows of the same type in all the sampling moments to obtain the deviation value corresponding to each time window.
7. The method of claim 1, wherein determining the performance of the sensor based on the deviation value corresponding to each time window comprises:
adding the deviation values corresponding to each type of time window to obtain an evaluation score of the sensor performance; wherein the evaluation score is inversely related to the performance of the sensor.
8. A sensor performance evaluation apparatus, comprising:
the acquisition module is used for acquiring the first data actually acquired by the sensor at each sampling moment and the second data acquired in a simulation way;
the first determining module is used for configuring multiple types of time windows for each sampling moment, and respectively determining a first characteristic and a second characteristic corresponding to each time window configured for each sampling moment according to the first data and the second data in each time window of each sampling moment; wherein, the corresponding set width of each type of time window is different;
the second determining module is used for determining the deviation value of each time window in the sensor sampling process according to the first characteristic and the second characteristic corresponding to each time window configured at each sampling moment;
and the evaluation module is used for determining the performance score of the sensor according to the deviation value corresponding to each time window.
9. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the sensor performance assessment method of any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the sensor performance evaluation method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410057671.5A CN117870749A (en) | 2024-01-15 | 2024-01-15 | Sensor performance evaluation method, device, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410057671.5A CN117870749A (en) | 2024-01-15 | 2024-01-15 | Sensor performance evaluation method, device, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117870749A true CN117870749A (en) | 2024-04-12 |
Family
ID=90592770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410057671.5A Pending CN117870749A (en) | 2024-01-15 | 2024-01-15 | Sensor performance evaluation method, device, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117870749A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118896638A (en) * | 2024-08-05 | 2024-11-05 | 广东深莱特科技股份有限公司 | A performance detection method and system for 3D perspective drying sensor |
-
2024
- 2024-01-15 CN CN202410057671.5A patent/CN117870749A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118896638A (en) * | 2024-08-05 | 2024-11-05 | 广东深莱特科技股份有限公司 | A performance detection method and system for 3D perspective drying sensor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12007270B2 (en) | Status detection method and apparatus for load cell | |
| CN110728008B (en) | Method and device for determining expected service life of intelligent ammeter | |
| CN117870749A (en) | Sensor performance evaluation method, device, electronic equipment and medium | |
| CN115795920A (en) | Product reliability evaluation method and device based on multi-stress coupling acceleration model | |
| CN116430271A (en) | Online detection method for LED soft light bar, intelligent terminal and storage medium | |
| CN113761755B (en) | Accelerated life analysis method under temperature and humidity dual stress by considering cognitive uncertainty | |
| CN119132003A (en) | Gas alarm detection method, system, device and gas alarm | |
| CN113032998B (en) | Medical instrument life assessment method and device | |
| CN118896638A (en) | A performance detection method and system for 3D perspective drying sensor | |
| CN116705210B (en) | Construction method of battery cell aging model and battery cell full life cycle performance prediction method | |
| US20130191071A1 (en) | System and method for automatic modal parameter extraction in structural dynamics analysis | |
| CN115376612B (en) | Data evaluation method and device, electronic equipment and storage medium | |
| CN112326882A (en) | Air quality sensor processing method and device | |
| CN114546841B (en) | Software quality assessment method based on cloud computing | |
| CN116466256A (en) | Battery monitoring method and device, vehicle and storage medium | |
| CN115659271A (en) | Sensor abnormality detection method, model training method, system, device, and medium | |
| CN112783763B (en) | Software quality detection method and device, electronic equipment and storage medium | |
| CN114254516A (en) | Parameter probability uncertainty modeling method under deleted data | |
| CN111272457A (en) | Mechanical state detection method based on temperature data and electronic equipment | |
| CN112241343A (en) | Slow disk detection method and device, electronic equipment and readable storage medium | |
| CN119005070B (en) | Method, device and medium for evaluating uncertainty of flow field parameters of high-speed wind tunnel test | |
| CN116307376A (en) | Method and device for obtaining carbon content of coal unit heat value | |
| CN110411657A (en) | A kind of intelligent pressure transmitter verification system | |
| CN118734611B (en) | Step stress test method and system for electric energy meter under combined action of temperature, humidity and stress | |
| CN116878801B (en) | Structure state identification method, device and terminal under random excitation effect |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |