WO2020021344A1 - 环境传感器协同校准方法 - Google Patents

环境传感器协同校准方法 Download PDF

Info

Publication number
WO2020021344A1
WO2020021344A1 PCT/IB2019/051244 IB2019051244W WO2020021344A1 WO 2020021344 A1 WO2020021344 A1 WO 2020021344A1 IB 2019051244 W IB2019051244 W IB 2019051244W WO 2020021344 A1 WO2020021344 A1 WO 2020021344A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
data set
calibration
sensor
station
Prior art date
Application number
PCT/IB2019/051244
Other languages
English (en)
French (fr)
Inventor
司书春
许军
秀福 万
帅帅 贾
一平 刘
Original Assignee
山东诺方电子科技有限公司
司书春
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Priority claimed from PCT/IB2018/055531 external-priority patent/WO2019150182A1/zh
Application filed by 山东诺方电子科技有限公司, 司书春 filed Critical 山东诺方电子科技有限公司
Priority to CN201980006118.6A priority Critical patent/CN112567241A/zh
Priority to PCT/CN2019/102420 priority patent/WO2020043031A1/zh
Priority to PCT/CN2019/102419 priority patent/WO2020043030A1/zh
Priority to CN201980089854.2A priority patent/CN113330283B/zh
Priority to CN201980089835.XA priority patent/CN113728220B/zh
Publication of WO2020021344A1 publication Critical patent/WO2020021344A1/zh

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00

Definitions

  • the present invention relates to a method for collaborative calibration of environmental sensors, and belongs to the field of environmental monitoring.
  • the monitoring indicators of atmospheric pollutants in environmental monitoring are sulfur dioxide, nitrogen oxides, ozone, carbon monoxide, PM1 (particles with aerodynamic particle size smaller than 1 micron), PM2.5 (aerodynamic particle size with smaller than 2.5 micron) in the atmosphere Particles), PM10 (particles with aerodynamic particle size less than 10 microns), PM100 (particles with aerodynamic particle size less than 100 microns), and VOCs (volatile organic compounds) or TVOC (total volatile organic compounds).
  • the atmospheric environment monitoring system can collect and process the monitored data, and timely and accurately reflect the regional ambient air quality status and changes.
  • the atmospheric environment monitoring equipment mainly includes fixed monitoring stations and mobile monitoring equipment.
  • the current fixed monitoring stations are mainly divided into large fixed monitoring stations and small stations.
  • Mobile monitoring equipment mainly includes special atmospheric environmental monitoring vehicles, drones, and handheld devices.
  • the above-mentioned small monitoring stations and handheld devices use air quality sensors to measure atmospheric pollutants.
  • Small sensors have the characteristics of low cost, miniaturization, and online monitoring, and can be used on a large scale.
  • the air quality sensor itself may cause errors due to inconsistencies between the measured values and the true values for various reasons.
  • air quality sensors Compared with large precision instruments or manual monitoring methods, air quality sensors have the characteristics of lower accuracy, poor stability, large errors, and frequent calibration.
  • the laser scattering method for air pollution particulate matter sensors has a broad market prospect because of its low cost and portability.
  • the portable analysis device using the scattering method has disadvantages such as poor measurement consistency, high noise, and low measurement accuracy.
  • the core device is easily affected by various environmental factors, and fluctuations easily cause misjudgment.
  • the sensor data changes suddenly and sharply being able to intelligently determine whether the change is due to a sensor failure or sudden pollution will greatly improve the reliability of the data and is of great value for ensuring the quality of environmental monitoring data.
  • Y data set the monitoring data of the mobile monitoring station
  • Fixed monitoring station A station with the ability to monitor the atmospheric environment. It can be a national control station, a calibration station, or a grid-based microstation.
  • Mobile monitoring station a monitoring station equipped with atmospheric environmental monitoring equipment and capable of moving. It can be a social vehicle equipped with miniature monitoring equipment, or it can be a professional atmospheric environment monitoring vehicle.
  • Contrast coefficient A quantity that indicates the degree of linear correlation between variables. It is generally represented by the letter tl.
  • the calibration coefficient in the present invention refers to a correction coefficient used for calibrating and correcting a deviation of a data set of a sensor. Particulate matter measured by light scattering method is easily affected by environmental factors, such as humidity and other factors. There are also many ways to improve sensor accuracy.
  • the current monitoring station calibration method mainly uses manual maintenance on a regular basis. Staff go to the site to clean and maintain the equipment, and carry standard equipment and standard gas to manually calibrate the sensors on site. Or simply make coefficient corrections to the monitoring equipment. These calibration methods have different levels of inaccuracy, complicated calibration and high cost. To address the above-mentioned shortcomings, the present invention provides a method for coordinated calibration of environmental sensors. Multiple sensors are used to calibrate and compare environmental sensors to achieve complementary data deviations.
  • Calibration method proposed by the present invention relates from (X stations of a data set from the data set p p monitoring stations and Y in the Y data set from the stations; first screened as a calibration reference based on a data set; This can be a subset of the a dataset; it can also be a subset of the (3 dataset or a subset of the Y dataset.
  • the benchmark data set As a benchmark data set, certain conditions must be met. First of all, when the benchmark data set is used to calibrate other monitoring stations with some distance, its data should be stable for a period of time without significant data fluctuations. Due to the natural diffusion of the atmosphere, this kind of data is stable for a period of time. It is the stability of air quality within a certain range, and the benchmark data set selected at this time should be able to represent data within a certain range.
  • the first calibration method proposed by the present invention is to calibrate the radon data set and the Y data set based on the a data set.
  • the data of the a data set are analyzed to determine the benchmark a data set.
  • Analyze data Set methods include direct average method, average method after removing the highest and lowest values, Kalman filter, Bayesian estimation, DS evidence reasoning, artificial neural network and other methods.
  • a calibration coefficient of the radon data set is obtained by comparing the radon data set with the benchmark a data set, which is used to calibrate the P data set.
  • a calibration coefficient of the Y dataset is obtained, which is used to calibrate the Y dataset.
  • the comparison method can be linear calibration, non-linear calibration, or other calibration methods.
  • multiple calibration coefficients are generally calculated, and calibration coefficients whose coefficients differ by less than a certain value are taken as valid calibration coefficients, and the average of these valid calibration coefficients is used as the final calibration coefficient to calibrate the calibration object.
  • the calibration coefficient may also need to take into account the spatial distribution.
  • the calibration coefficients of (3 datasets can be weighted according to the distance from (3 sites to a site), the closer the distance is, the greater the weight; if the opposite site is within a certain distance from a site, the weighted average is accurate as the calibration target.
  • Value Take the data within a certain distance from Site A to Site Y as valid data to participate in the calibration calculation.
  • the calibration coefficients of the Y data set can also be determined according to different data intervals of the data, that is, multiple calibration coefficients are set in different data intervals. In the selection of calibration coefficients in different intervals, direct average method, average method after removing the highest value and the lowest value can still be used.
  • the calibration P data set is based on the Y data set, and the calibration (3 data set. When the distance between the mobile monitoring station and the fixed monitoring station is less than the calibration trigger distance t, the Y data set of the mobile monitoring station is used as the reference Compare the 3 data set of the fixed monitoring station with the Y data set of the mobile monitoring station to obtain a calibration coefficient, and use the calibration coefficient and other calibration factors to calibrate the fixed station.
  • the calibration method can choose linear calibration or non-linear calibration, and the calibration trigger
  • the distance t can be 500m, 1km, 2km, 5km, etc.
  • calibrate other Y data sets Based on the selected Y data set, calibrate other Y data sets.
  • the Y data set of the selected mobile monitoring station is compared with the Y data set of other mobile monitoring stations based on the Y data set of the selected mobile monitoring station to obtain Calibration coefficient, use fixed calibration coefficient and other calibration factors to calibrate fixed stations.
  • Calibration method can choose linear calibration or non-linear calibration, calibration trigger distance t Be a 500m, lkm s 2km s 5km equidistant calibration third embodiment the present invention set forth Ranking calibration data set to the P and Y to the credibility of the data set ranked by confidence to high confidence device Low equipment calibration.
  • the reliability can be the comparison coefficient between the reference data and the number being calibrated, or other parameters of the monitoring equipment such as the calibration time factor (representing the time since the last calibration), the stability factor, and so on.
  • the P data set and the Y data set are compared with the a data set to obtain a credibility index.
  • the comparison method can be a correlation coefficient, a proportional average, and the like.
  • the reliability index is ranked from high to low, and the lower-ranked data set is calibrated.
  • the first method is used for the calibration method. Recalculate the confidence level after calibration to rank. For the P fixed station, select a national control station within a certain range for credibility calculation.
  • the credibility can be the average of multiple national control stations, or it can be calculated by weighted average based on the distance as a weight; in the case where there is no national control station within a certain range Calculate the credibility with the average value of the a dataset for the entire city.
  • the reliability calculation is performed on the data after mobile station Y moves to a certain range of national control station a. After ranking, the higher-ranked data set is compared with the lower-ranked data set, the calibration coefficient is calculated, and the higher-ranked data set is used to calibrate the lower-ranked data set.
  • the monitoring data of a fixed monitoring station When the monitoring data of a fixed monitoring station is abnormal, it can communicate with the mobile monitoring station to control the working state of the sensors of the mobile station, and increase the monitoring frequency and data return frequency.
  • the abnormal data can be judged by the contrast coefficient exceeding the setting range, that is, the abnormal data of the station is determined.
  • x is the calibrated data
  • y c is the corrected reference data
  • ri is the contrast coefficient
  • the calibration coefficient is the median of the contrast coefficient.
  • the calibration coefficient is the mode of the contrast coefficient.
  • the calibration coefficient is the value of the comparison coefficient calculated by other probability methods.
  • the distance factor between the reference station and the calibrated station can also be considered, and a distance factor is introduced.
  • the distance factor / d is used to consider the influence of the distance factor between the monitored point and the monitoring station on the reliability of the monitoring data.
  • the distance factor can be normalized by the inverse ratio of the distance from the geometric center point of the area to each monitoring point.
  • the distance factor can be obtained from the data obtained by monitoring stations in a certain area.
  • the pollution data consists of monitoring data from several nearby monitoring stations. These monitoring data can have different weights for pollution data at a specific location.
  • the weight is normalized by the inverse ratio of the distance from the specific location to each monitoring point.
  • the weight is the distance factor.
  • d represents the distance from the geometric center point of the area to each station in the area or the distance from the specific position to each monitoring point
  • the set value of the distance is represented by A.
  • the distance factor is 1. After exceeding the set distance A, the farther the distance is, the smaller the weight of the monitoring station data is, and the closer the distance is, the greater the weight of the monitoring station data is.
  • the formula for calculating the distance factor is:, d> A 0 ⁇ d ⁇ A d represents the distance from the geometric center point of the area to each station in the area, or from the specific location to each monitoring point.
  • the parameter K is a distance weight parameter.
  • the reference value data can also be obtained through normalization. Apply the normalized base station correction calculation formula to correct the base station data.
  • the normalized calculation formula: y c is the revised baseline data,
  • n is the number of reference stations that meet the standard. In the case where there is only one reference station that meets the standard, the correction of the reference data is calculated as follows:
  • the stability coefficient 2 is the ratio of the number of base station data in the set interval to the total base station data. If 2 is greater than the set percentage (the set percentage can be 80%, 90%, and other percentages), the base station data set is considered stable. A higher A indicates a more stable data set.
  • the setting interval is the range given to the reference data within the setting T time range. The mathematical expression of the setting interval is
  • Y can be obtained from statistical methods such as the average, median, and mode of the base station data within the time range of T, "" is the interval coefficient.
  • the number of base station data that falls within the set interval within the T time range is the number of base station data that falls within the set interval within the T time range
  • the stability coefficient can also be related to the variance of the reference data in the set time range. , Then it is unstable. If the base is set within the T time range, according to the variance ⁇ Variance setting, B, then it is stable.
  • C) The stability factor can also be related to the standard deviation of the reference data within the set 1T time range. Set the standard deviation of the reference data within the T time range, and the variance is set to ⁇ C, then it is unstable. If the standard deviation of the reference data within the T time range is set to ⁇ Variance setting, ⁇ C, it is stable. For mobile monitoring stations to be used as a calibration reference, they need to have sufficient credibility.
  • the low-frequency sensor disclosed in the earlier application PCT / IB2018 / 05531 discloses air pollution detection equipment.
  • the air pollution detection equipment that is, the mobile monitoring station in this article, includes a main control module and a detection module.
  • the detection module uses at least four The sub-sensor units constitute a sensor module; when the main control module detects a suspected abnormality in one of the sub-sensor units, and judges that the suspected abnormal sensor is an abnormal sensor, isolates the abnormal sensor, and the abnormal sensor In the isolation zone, the multi-core sensor module continues to work normally after it is downgraded.
  • the air pollution detection device includes a main control module and a detection module; the detection module includes at least two similar sub-sensor units to form a sensor module; and the sub-sensor units work At normal operating frequency.
  • the detection module further includes at least one sub-sensor unit similar to the sensor module to form a low-frequency calibration module; the sub-sensor unit in the low-frequency calibration module operates at a frequency much lower than the operating frequency of the sensor module. Therefore, the low-frequency calibration module is also called a low-frequency group.
  • the sensor module is also called a high-frequency group.
  • the operating frequency of the sensor module is 10 times or more than that of the low-frequency calibration module.
  • the ratio of the working frequency of the high-frequency group to the low-frequency group is called the high-frequency and low-frequency ratio, and can be selected as: 2: 1, 3: 1, 4: 1, 5: 1, 6: 1, 7: 1, 8: 1 , 9: 1, 10: 1, 15: 1, 20: l o
  • the working frequency of the low frequency group can be consistent with the rhythm of abnormal judgment. That is, when it is necessary to determine whether there is an abnormal phenomenon of the sub-sensor in the sensor module, the low-frequency group performs the detection work.
  • the accuracy of its data can be restored by calibration; that is, the sub-sensor that is not attenuated or has a very low attenuation is used to calibrate the high attenuation.
  • Child sensor During the operation of the sensor module, every certain period of time, such as 1 day, 1 week, or 1 month, use the low-frequency group detection data as a reference, calibrate the high-frequency group detection data, and the calibration coefficients can use the high-frequency group sensor detection data. The ratio of the average value to the average value of the detection data of the low frequency group.
  • Isolation and Recovery Prior application PCT / IB2018 / 05531 also discloses a set of methods to identify the working status of sub-sensors and isolate and restore the sub-sensors.
  • the sensor module obtains a set of detection data at a time, and the main control module filters out data that is suspected to be abnormal from this set of data, and then determines whether the corresponding sub-sensor meets the isolation condition. After determining that the sub-sensor is an abnormal sub-sensor, the abnormal sub-sensor is classified into the isolation area; after determining that the sub-sensor suspected to be abnormal does not satisfy the isolation condition, the sub-sensor continues to work normally. It is judged whether the sub-sensor entering the isolation zone can heal itself.
  • the frequency reduction work processing is performed on the self-healing sub-sensor. Sub-sensors that cannot heal themselves will stop working and notify the operation and maintenance party for repair or replacement.
  • the main control module detects the output data of the sub-sensor to determine whether it has reached the recovery condition. The sub-sensor that has reached the recovery condition is detached from the isolation area and resumes work. The output data participates in sensor module data or the main control Data calculation; judge again whether the abnormal sensor that does not meet the recovery conditions can heal itself. After isolating the abnormal sub-sensors in the sensor module, the average value of the remaining sub-sensor output data is used as the output result of the sensor module, and the sensor module can continue to be used normally.
  • FIG. 1 is a schematic diagram of a calibration system
  • FIG. 2 is a schematic diagram of a calibration data set and a Y data set based on a data set
  • FIG. 3 is a schematic diagram of a calibration data set based on a data set
  • FIG. 4 It is a schematic diagram of the calibration area range.
  • Fig. 5 is a flowchart of calibrating the radon data set and the Y data set based on the a data set
  • Fig. 6 is a flowchart of calibrating the radon data set based on the Y data set
  • the Y data set is used as a basis to calibrate the Y data set.
  • Figure 8 is the flowchart for ranking calibration.
  • 10 is the reference station
  • 20 is the fixed monitoring station
  • 30 is the mobile monitoring station
  • 40 is the data center
  • 50 is
  • 101 is the No. 1 base station
  • 102 is the No. 2 base station
  • 103 is the No. 3 base station
  • 201 is the No. 1 fixed calibration station
  • 202 is the No. 2 fixed calibration station
  • 203 is the No. 3 fixed calibration station.
  • Calibration station, 301 is No. 1 mobile calibrated station
  • 302 is No. 2 mobile calibrated station
  • 501 is the calibration range of the base station.
  • the calibrated stations are sorted according to their credibility and used The most accurate micro-station data is calibrated one by one to get the calibration coefficients, and the above data is counted in the following table:
  • the spatial distribution may also need to be considered during calibration.
  • the station to be calibrated is ranked lower.
  • P 4 150
  • the contrast coefficient is within the range of 0.95-1.05 with high credibility, and no calibration is performed; the contrast coefficient is between 1.05-1.2, and calibration is performed; the contrast coefficient is greater than 1.2 to not perform calibration, and the equipment may have a serious failure.
  • the control system will be alerted to indicate that the monitoring equipment needs manual maintenance.
  • the calibration range is determined by the correlation coefficient, and the equipment with the correlation coefficient greater than 0.9 will not be calibrated. For the equipment with the correlation coefficient less than 0.9, the calibration will be performed with the goal of reaching the benchmark data set.
  • the device with the highest reliability is used for calibration As a benchmark, if the device with the highest credibility is a fixed station, calibration is performed from the stations around the fixed station until all is completed. If the device with the highest credibility is a Y mobile station, the The stations passing by it are calibrated as priority calibration objects until they are all completed.
  • the calibration range is determined by the proportional average coefficient.
  • Equipment with a proportional coefficient of 0.9 to 1.1 is not calibrated, and equipment with a proportional coefficient in other ranges is used to reach the benchmark.
  • the data set is calibrated for the target.
  • the device with the highest credibility is used as the calibration benchmark. If the device with the highest credibility is a fixed station, the calibration is started from the stations around the fixed station. Until all is completed, if the device with the highest credibility is the Y mobile station, the stations passing by around it are used as priority calibration objects for calibration until all are completed.
  • Embodiment 5 As shown in FIG. 2, there are reference stations No. 1, 2, and 3 in the area, and two calibrated stations (31, yl. Take four state-controlled reference stations at four moments, Tl, T2, T3, and T4. And the monitoring values of the fixed micro-stations, the average values of the monitoring data of each fixed national control base station are calculated as shown in the table below. 1.04.
  • the calibration coefficient can also be determined according to different data intervals of the data, that is, multiple calibration coefficients are set in different data intervals. In the selection of calibration coefficients in different intervals, the direct average method can still be used, after removing the highest and lowest values Methods such as averaging method. In this embodiment, it is specified that the contrast coefficient is within the range of 1-1.2, and the calibration coefficient is the average value of the contrast coefficient between 1-1.2; if the contrast coefficient is above 1.2, the maximum value of the contrast coefficient is removed. Take the average.
  • the seventh embodiment provides that the calibration procedure is started when the mobile monitoring equipment enters a range of 5 km around the reference station.
  • vehicle No. 1 is located within a 5 km range around No. 1, 2, and 3 monitoring stations, and No. 2 If the vehicle is not within a 5km area around monitoring stations 1, 2, and 3, the mobile monitoring device No. 1 starts the calibration process, and the mobile testing device No. 2 does not start the calibration process.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Food Science & Technology (AREA)
  • Medicinal Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

一种环境传感器协同校准的方法,涉及环境监测领域。该方法涉及来自于α监测站的α数据集、来自于β监测站的β数据集和来自于γ监测站的γ数据集,监测站的环境传感器间采用多数据相互校准比对的方式,实现数据偏差互补,相互校验,提高了传感器的可靠性、一致性以及精度。

Description

环境传感器协同校准方法 技术领域 本发明涉及一种环境传感器协同校准的方法, 属于环境监测领域。 背景技术 环境监测中大气污染物监测指标为大气中的二氧化硫、氮氧化物、臭氧、一氧化碳、 PM1(空 气动力学粒径小于 1微米的粒子)、 PM2.5(空气动力学粒径小于 2.5微米的粒子)、 PM10(空 气动力学粒径小于 10微米的粒子)、 PM100(空气动力学粒径小于 100微米的粒子)和 VOCs (挥发性有机物) 或 TVOC (总挥发性有机物) 。 大气环境监测***可以对监测的数据进行 收集和处理, 并及时准确地反映区域环境空气质量状况及变化规律。 现在的大气环境监测设备主要有固定式监测站和移动式监测设备。目前的固定式监测站主要 分为大型固定监测站点和小型站点。移动式监测设备主要有专用大气环境监测车、 无人机以 及手持设备等。上述小型监测站点、手持设备都用到了空气质量传感器来测量大气中的污染 物。 小型传感器具有低成本、 小型化和在线监测的特点, 可以大规模使用。 空气质量传感器 本身会由于各种原因造成测得值与真实值不一致而存在误差。与大型精密仪器或者手工监测 方式相比, 空气质量传感器还有精确度更低、 稳定性差、 误差大、 需要经常校准的特点。 激光散射法的大气污染颗粒物传感器, 因为低成本和便携性, 有着宽广的市场前景。 然而采 用散射法的便携式分析装置就会存在测量一致性差、 噪声大、 测量精度低等缺点, 核心器件 容易受到各种环境因素影响, 而波动容易引起误判。 当传感器数据突然大幅变化时, 能够智能判断出变化原因是传感器故障还是突发污染, 将会 极大提高数据可靠性, 对于保证环保监测数据质量具有重要价值。 当设备发生故障时, 如果 能够通过自动修复, 也可以大幅提高数据的在线率, 对于治霾工作所需的连续监测具有重要 价值。 同时又可以节省设备维护保养方面的人力物力, 减少社会浪费。
发明内容
术语解释 a数据集:基准站的 (国控站、 市控站、 单独设置的校准站) 的监测数据; al表示基准站在 T=1时刻的数据或者数据组; A1表示一个国控站在 T=1时刻的数据。
P数据集:固定式监测站的监测数据, (31表示固定大气网格化微站在 T=1时刻的数据或监测 数据组; B1表示固定大气网格化微站在 T=1时刻的数据。
Y数据集:移动式监测站的监测数据, Yl表示固定大气网格化微站在 T=1时刻的数据或监测 数据组; Y1表示固定大气网格化微站在 T=1时刻的数据。 固定监测站:具备大气环境监测能力的站点, 可以是国控站、 校准站、 网格化微站。 移动监测站:搭载大气环境监测设备, 并具备移动能力的监测站点。 可以是搭载了微型监测 设备的社会车辆, 也可以是专业的大气环境监测车辆。 对比系数:表示变量之间线性相关程度的量, 一般用字母 tl表示。 校准系数:校准系数在本发明中指在用于校准、 修正传感器的数据集偏差的修正系数。 光散射法测量颗粒物易受环境因素影响测量精度, 如湿度等因素。 目前也出现了多种提高传 感器精度的方式。 目前的监测站校准方式主要采用定期人工维护, 工作人员到现场对设备进行清理维护, 并携 带标准设备和标气, 对传感器进行现场的手工校准。 或者简单对监测设备进行系数修正。 这 些校准方式不同程度的存在依然不精确、 校准复杂和成本高的问题 针对上述不足, 本发明提供了环境传感器协同校准的方, 环境传感器间采用多数据相互校准 比对的方式, 实现数据偏差互补, 相互校验, 提高传感器的可靠性 致性、精度以及寿命。 本发明提出的校准方法, 涉及来自于(X监测站的 a数据集、 来自于 p监测站的 p数据集和来自 于 Y监测站的 Y数据集;首先筛选出作为校准依据的基准数据集;这个可以是 a数据集的子 集;也可以是(3数据集的子集或者 Y数据集的子集。
具体步骤为 :
1) 首先获取 a数据集、 卩数据集、 Y数据集中至少两种;
2) 筛选出作为校准依据的基准数据集, 以及被校准数据集;
3) 从基准数据集得到基准数据 (y), 从被校准数据集得到被校准数据 (x);
4) 依据基准数据 (y) 和被校准数据 (x) 得到对比系数 (11);
5) 计算校准系数 (c); 被校准监测站采用校准系数 (c) 进行校准。
作为基准数据集, 必须满足一定的条件。 首先, 基准数据集在用于校准有一些距离的其他监测站时, 其数据应当在一段时间内是比较 稳定的, 没有显著的数据波动;由于大气的自然扩散, 这种一段时间的数据稳定代表的是一 定范围内的空气质量的稳定, 此时选出的基准数据集, 应当能够代表一定范围内的数据。 我们用一个数据集的子集的稳定系数来表征这个子集中数据的稳定性。在不是近距离校准的 情形下, 只有稳定系数达到预期值的数据集的子集才能被选为基准数据集。 以 cx数据集为依据, 校准 P数据集和 Y数据集 本发明提出的第一种校准方式为以 a数据集为依据, 校准卩数据集和 Y数据集。 在卩数据集和 Y 数据集达成均匀度要求的情况下, 通过分析 a数据集的数据, 确定基准 a数据集。 分析 a数据 集的方法有直接平均值法、 去掉最高值和最低值后平均法、 卡尔曼滤波、 贝叶斯估计、 D-S 证据推理、 人工神经网络等方法。 确定基准 a数据集后, 通过将卩数据集与基准 a数据集作比较, 得出卩数据集的校准系数, 用于 校准 P数据集。 同理, 通过 Y数据集与基准 a数据集作比较, 得出 Y数据集的校准系数, 用于校 准 Y数据集。 比较的方式可以采用线性校准的方式, 也可以采用非线性校准以及其他校准方 式。 校准过程中, 一般计算多个校准系数, 取系数相差小于一定值的校准系数为有效校准系数, 将这些有效校准系数的平均值作为最终的校准系数, 对校准对象进行校准。 校准系数还可以需要考虑空间分布。 对(3数据集的校准系数可以根据(3站点距离 a站点的距离 做权重排序, 距离越近权重越大;对卩站点距离 a站点一定距离以内的情况下, 取加权平均值 为校准目标准确值。 对 Y站点取经过 a站点一定距离内的数据为有效的数据参与校准计算。
Y数据集的校准系数还可以根据数据不同数据区间而确定, 即在不同数据区间设定多个校准 系数。在不同区间的校准系数选择上仍然可以使用直接平均值法、 .去掉最高值和最低值后平 均法等方法。 以 Y数据集为依据, 校准 P数据集 以 Y数据集为依据, 校准(3数据集。 当移动监测站与固定监测站距离小于校准触发距离 t时, 以移动监测站的 Y数据集为基准,将固定监测站的(3数据集与移动监测站的 Y数据集进行比对, 得到校准系数, 利用校准系数及其他校准因子校准固定站点。校准方法可以选择线性校准或 者非线性校准, 校准触发距离 t可以是 500m、 1km、 2km、 5km等距离。 以选定 Y数据集为依据, 校准其他 Y数据集 以选定的 Y数据集为依据, 校准其他 Y数据集。 当选定移动监测站与其他移动监测站距离小于 校准触发距离 t时, 以选定移动监测站的 Y数据集为基准, 将选定移动监测站的 Y数据集与其 他移动监测站的 Y数据集进行比对, 得到校准系数, 利用校准系数及其他校准因子校准固定 站点。 校准方法可以选择线性校准或者非线性校准, 校准触发距离 t可以是 500m、 lkm s 2km s 5km等距离。 排名校准 本发明提出的第三种校准方式为将 P数据集和 Y数据集以可信度排名后, 由可信度高的设备向 可信度低的设备校准。 可信度可以是基准数据与被校准数的对比系数,还可以是监测设备的其他参数如校准时间因 子 (表示距离上次校准的时间)、 稳定系数等。
P数据集和 Y数据集排名示意表
Figure imgf000006_0002
P数据集和 Y数据集通过和 a数据集进行对比,得到可信度指标,对比的方式可以是相关系数、 比例均值等方式。 得到可信度指标后, 将可信度由高到低进行排名, 对排名靠后的数据集进 行校准, 校准方式采用第一种方法。 校准后重新计算可信度进行排名。 对于 P固定站, 选取其一定范围内的 a国控站进行可信度计算。在一定范围内存在多个国控站 的情况下, 可信度可以是多个国控站的均值, 也可以根据距离作为权重进行加权平均进行计 算;在一定范围内没有国控站的情况下, 以整个城市的 a数据集的均值进行可信度计算。 对 Y移动站, 当 Y移动站移动至 a国控站一定范围后的数据进行可信度计算。 排名后, 排名较高 的数据集与排名较低的数据集进行比对, 计算校准系数, 利用排名较高的数据集校准排名较 低的数据集。 监测站协同工作的方法 当固定监测站的监测数据异常时, 可以与移动监测站进行通信, 控制移动站的传感器的工作 状态, 提高监测频率和数据回传频率。 数据异常的判定可以是对比系数超出设定范围, 即判 定该站点数据异常。 相关计算方法 比对公式:
X
T1 =—
yc
x为被校准数据, yc为经过修正后的基准数据, ri为对比系数。
校准公式:
Figure imgf000006_0001
X’ 为校准后数据, C为校准系数, C可以由 T1的平均值得到或者 T1经过其他数学运算得到。 其 他数学运算可以是去掉最高值和最低值后平均法、卡尔曼滤波、贝叶斯估计、 D-S证据推理、 人工神经网络等方法。 获得校准系数 C的方法: a) 校准系数是对比系数的平均值, c = fi
b) 校准系数是对比系数的中位数 c) 校准系数是对比系数的众数 d) 校准系数是对比系数经过其他概率方法计算得出的值。 在校准过程中还可以考虑基准站与被校准站之间的距离因素, 引入距离因子。其中距离因子 /d用于考量被监测点与该监测站之间的距离因素所产生的对监测数据可靠性的影响。距离因 子可以由某一区域内监测站点所获取数据占据该区域的权重由该区域几何中心点到各个监 测点的距离的反比归一化得到;距离因子另一种体现方式是, 某一特定位置的污染数据由相 近的数个监测站点的监测数据组成,这些监测数据对特定位置的污染数据可以有不同的权重, 该权重由该特定位置到各个监测点的距离的反比归一化得到, 该权重就是距离因子。 在 计算中, d 表示该区域几何中心点到该区域内各个站点之间的距离或该特定位置到各 个监测点的距离, 该距离的设定值用 A表示。 在设定距离 A以内, 距离因子为 1 ;超过设定 距离 A后, 距离越远则该监测站点数据所占据的权重越小, 距离越近则该监测站点数据所占 据的权重越大。 距离因子计算公式为 : , d > A
Figure imgf000007_0001
0 < d < A d 表示该区域几何中心点到该区域内各个站点之间的距离或该特定位置到各个监测点的距 禹。 参数 K是距离权重参数, 距离权重越小, 代表距离的影响越大, 一般情况下 K = A 应用距离因子后, 基准值数据还可以通过归一化运算得到。应用归一化法基准站修正计算公 式修正基准站数据, 归一化计算公式:
Figure imgf000007_0002
yc为经过修正后的基准数据,
y'为未经修正的基准站数据,
n为达到标准的基准站个数。 在仅有一个达到标准的基准站的情况下, 基准数据的修正计算方式如下:
yc = y - fd x (x - y) 稳定系数 A的获得方法: a) 稳定系数 2为设定区间内的基准站数据数量占总基准站数据数量的比値。 若 2大于设定百分 比(设定百分比可以是 80%、 90%等其他百分比),则认为该基准站数据集稳定, A越高代 表数据集越稳定。 设定区间为设定 T 时间范围内的给予基准数据的范围, 设定区间的数学表示为
(y - U X y , y + U X y) , Y可以由 T时间范围内基准站数据的平均值、 中位数、 众数等 统计方法得来, "为区间系数。
T时间范围内落入设定区间内的基准站数据的数量
Figure imgf000008_0001
T时间范围内基准站数据的数量 b) .稳定系数还可以与设定区间为设定 时间范围内的基准数据的方差有关 如果设定 T时间范围内的基》据方差 >方差设定値 S3 ,则不稳定 如果设定 T时间范围内的基®据方差 <方差设定値 B ,则稳定 C) 稳定系数还可以与设定区间为设定 1T时间范围内的基准数据的标准差有关 如果设定 T时间范围内的基准数据的标准差 ñ方差设走値 C ,则不稳定 如果设走 T时间范围内的基准数据的标准差 <方差设定値 C ,则稳定 对于移动监测站来说, 用来作为校准基准时, 需要其具备足够的可信度。 当移动监测站采用冗佘多传感器设计时, 移动监测站的可信度会得到大幅度提高。 髙低频传感器 在先申请 PCT/IB2018/05531中公开了大气污染检测设备, 所述大气污染检测设 备, 也就是本文中的移动监测站, 包含主控模块和检测模块; 所述检测模块采用至 少四个子传感器单元组成传感器模组; 当主控模块发现其中一个子传感器单元出现 疑似异常, 并判断所述疑似异常子传感器为异常子传感器后, 对所述异常子传感器 进行隔离, 所述异常子传感器归入隔离区, 多核传感器模组降级后继续正常工作。 本申请进一步公开了另一种大气污染检测设备, 所述大气污染检测设备包含主控模 块和检测模块; 所述检测模块包含至少两个同类子传感器单元组成传感器模组; 所 述子传感器单元工作在正常的工作频率。 所述检测模块还包含至少一个与传感器模 组同类的子传感器单元组成低频校准模组; 低频校准模组内的子传感器单元工作在 远低于传感器模组的工作频率。 因此低频校准模组也称之为低频组。 作为对照, 传 感器模组也称之为高频组。 通常, 传感器模组的工作频率是低频校准模组的 10倍或以上。 高频组和低频组的 工作频率的比率, 称为高频低频比, 可以选择为: 2: 1, 3 : 1, 4: 1, 5: 1, 6: 1, 7: 1, 8: 1, 9: 1, 10: 1, 15: 1, 20: l o 低频组的工作频率可以与异常判断的节奏保持一致。 也就是说, 当需要对传感器模 组中是否存在子传感器异常现象进行判断时, 低频组才进行检测工作。 由于激光功率衰减在激光传感器的工作寿命内的大多数时间是缓慢进行的, 是可以 通过校准来恢复其数据的准确性; 也就是使用未衰减或衰减程度非常低的子传感器 来校准衰减程度高的子传感器。 在传感器模组运行过程中, 每隔一定时间, 例如 1天, 1周或 1个月, 使用低频组 检测数据作参考, 校准高频组检测数据, 校准系数可以使用高频组传感器的检测数 据平均值与低频组检测数据平均值之比得到。 除了激光传感器的光衰效应, 其他类型的传感器, 也存在长时间高负荷工作情况下 的性能不稳定或者数据误差增大的可能倾向。 通过引入一个低频组, 能够作为相对 可靠的基准, 用来判断传感器模组是否存在数据偏移现象。 同时, 由于低频组的数据通常可信度更高, 在判断传感器模组中哪个子传感器单元 属于疑似异常或异常时, 可以通过增加低频组的数据权重, 来做出更可信的判断。 一种简单的方案是所有的低频组数据按两倍权重参与疑似异常判断。 隔离与恢复 在先申请 PCT/IB2018/05531还公开了一套识别子传感器工作状态并对子传感器进 行隔离和恢复的方法。 传感器模组获得一个时刻的一组检测数据, 主控模块从这一 组数据中筛选出疑似异常的数据, 进而判断相应的子传感器是否满足隔离条件。 判 断子传感器为异常子传感器后将异常子传感器归入隔离区; 判断疑似异常的子传感 器不满足隔离条件后, 该子传感器继续正常工作。 判断进入隔离区的子传感器是否 可以自愈, 如果判断可以自愈则对该可自愈的子传感器做降频工作处理, 但是子传 感器输出的数据不参与主控模块输出数据的计算。 对于无法自愈的子传感器则停止 工作, 并通知运行维护方进行维修或者更换。 对于降频后的子传感器, 由主控模块 检测其输出的数据, 判断其是否达到恢复条件, 将达到恢复条件的子传感器调离隔 离区, 恢复工作, 输出数据参与传感器模组数据或主控数据计算; 对于不符合恢复 条件的异常子传感器再次进行是否可自愈的判断。 将传感器模组中异常子传感器隔离后, 剩余的子传感器输出数据平均值作为传感器 模组的输出结果, 传感器模组可以继续正常使用。
附图说明 图 1为校准***的组成示意图; 图 2为以 a数据集为依据, 校准卩数据集和 Y数据集示意图; 图 3为以 a数据集为依据, 校准卩数据集示意图; 图 4为校准区域范围示意图; 图 5为以 a数据集为依据, 校准卩数据集和 Y数据集的流程图; 图 6为以 Y数据集为依据, 校准卩数据集的流程图; 图 7为以 Y数据集为依据, 校准 Y数据集的流程图; 图 8为排名校准的流程图; 图中: 10为基准站, 20为固定监测站, 30为移动监测站, 40为数据中心, 50为用户, 101 为 1号基准站、 102为 2号基准站、 103为 3号基准站、 201为 1号固定式被校准站、 202 为 2号固定式被校准站、 203为 3号固定式被校准站、 301为 1号移动式被校准站、 302为 2号移动式被校准站、 501为基准站校准范围。
具体实施方式 实施例 _ 如图 2所示, 在区域内存在 1、 2、 3号固定国控基准站, 一个固定式被校准站。取 Tl、 T2、 T3、 T4 四个时刻的四个国控基准站以及固定式微站的监测值, 统计各固定国控基准站的监 测数据平均值如下表,根据表中数据计算被校准微站与国控基准站的平均值的比值 n分别为 1.2、 1.1、 1.1、 1.04, 所以计算得到 n的平均值 = 1.11。 在本实施例的比较中使用的是算术 平均值计算三个国控基准站监测数据的平均值, 还可以使用归一化法进行计算。 同理, 可以 以 a数据集为依据, 校准 Y数据集得到 ri平均值
Figure imgf000011_0001
实施例二 如图 3所示, 已知图中固定的 1号被校准站、 2号被校准站、 3号被校准站的监测数据分别 为 =120、 (32=115、 p3=110, 基准站的数据为 a=110, 1、 2、 3号固定式微站距离国控基准 站的距离分别为 5km、 6km、 7km , 根据以上数据按照可信度对被校准站进行排序, 并使用 最准确微站数据逐一向下校准得到校准系数, 将上述数据统计至下表中 :
Figure imgf000012_0005
应用校准公式后, 经校准的 经校准的
Figure imgf000012_0001
Figure imgf000012_0002
实施例三 如图 3所示, 已知图中固定的 1号被校准站、 2号被校准站、 3号被校准站的监测数据分别 为 =120、 (32=115、 p3=110, 基准站的数据为 a=110, 1、 2、 3号固定式微站距离国控基准 站的距离分别为 5km、 6km、 7km。校准时还可以需要考虑空间分布。对卩数据集的校准系数 可以根据 P站点距离 a站点的距离彳故排序, 距离越近距离因子越大, 应用距离因子计算公式, 取 A=5 ;根据距离因子对被校准站进行排序, 并使用排名第一的被校准站校准排名靠后的被 校准站。
Figure imgf000012_0006
应用校准公式后, 经校准的 经校准的 为
Figure imgf000012_0004
Figure imgf000012_0003
实施例四 已知图中固定的 1号被校准站、 2号被校准站、 3号被校准站、 4号被校准站的监测数据分 别为 =120、 (32=115、 p3=110、 p4=150, 基准站的数据为 a=110, 根据以上数据按照可信度 对微站进行排序, 并使用最准确微站数据逐一向下校准得到校准系数, 将上述数据统计至示 意表 2中。规定对比系数在 0.95-1.05范围内的可信度较高, 不进行校准;对比系数在 1.05 - 1.2之间, 进行校准;对比系数大于 1.2以不进行校准, 设备可能发生较严重故障, 向控制系 统报警提示该监测设备需要人工维护。 校准范围的确定以相关系数确定, 相关系数大于 0.9的设备不进行校准, 对于相关系数小于 0.9 的设备, 则以达到基准数据集为目标进行校准, 原则上以可信度排名第一的设备为校准 基准, 若可信度排名第一的设备为卩固定站, 则从该(3固定站周围的站点开始进行校准, 直至 全部完成, 若可信度排名第一的设备为 Y移动站, 则以其周围经过的站点为优先校准对象进 行校准, 直至全部完成。 校准范围的确定以比例均值系数确定, 比例系数在 0.9〜 1.1的设备不进行校准, 比例系数在 其他范围的设备, 则以达到基准数据集为目标进行校准, 原则上以可信度排名第一的设备为 校准基准, 若可信度排名第一的设备为卩固定站, 则从该(3固定站周围的站点开始进行校准, 直至全部完成, 若可信度排名第一的设备为 Y移动站, 则以其周围经过的站点为优先校准对 象进行校准, 直至全部完成。
Figure imgf000013_0001
实施例五 如图 2所示, 在区域内存在 1、 2、 3号基准站, 两个被校准站(31、 yl。 取 Tl、 T2、 T3、 T4 四个时刻的四个国控基准站以及固定式微站的监测值,统计各固定国控基准站的监测数据平 均值如下表,根据表中数据计算被校准微站与国控基准站的平均值的比值 n分别为 1.2、1.1、 1.1、 1.04。校准系数还可以根据数据不同数据区间而确定, 即在不同数据区间设定多个校准 系数。在不同区间的校准系数选择上仍然可以使用直接平均值法、去掉最高值和最低值后平 均法等方法。 在本实施例中规定对比系数在 1-1.2区间内, 校准系数取对比系数在 1-1.2之 间对比系数的平均值;对比系数在 1.2以上的, 去掉对比系数最大值后取均值。
Figure imgf000013_0002
实施例六 已知某一时刻, 有 1、 2、 3、 4、 5号固定式国控基准站的监测数据分别为 Cd=120、 a2=115、 a3=110 s 0(4=150, a5=120, 计算区域平均值, 则区域平均值 =123, 取各站点数据与平均值 相差 ±10%以内的数据的平均值作为基准值用于校准。
Figure imgf000014_0001
实施例七 规定当移动式监测设备进入到基准站周围 5km的范围内开启校准程序, 如图 4所示, 1号车 辆处在 1、 2、 3号监测站周围 5km范围内的区域, 2号车辆没有处在 1、 2、 3号监测站周围 5km范围内的区域, 则 1号移动式监测设备启动校准过程, 2号移动式检测设备不启动校准 过程。

Claims

权利要求书
1 一种环境传感器协同校准的方法,所述方法涉及来自于 a监测站的 a数据集、 来自于卩监测站的卩数据集和来自于 Y监测站的 Y数据集;其步骤为 :
1) 首先获取 a数据集、 卩数据集、 Y数据集中至少两种;
2) 筛选出作为校准依据的基准数据集, 以及被校准数据集;
3) 从基准数据集得到基准数据(y) ,从被校准数据集得到被校准数据(x) ;
4) 依据基准数据 (y) 和被校准数据 (x) 得到对比系数 (n) ;
5) 计算校准系数 (c) ;被校准监测站采用校准系数 (c) 进行校准; 其特征在于, 所述基准数据集中的数据满足稳定系数 U) 的要求。
2. 如权利要求 1所述的方法,其特征在于,所述基准数据集是 a数据集的子集、
(B数据集的子集或者 Y数据集的子集;被校准数据集是卩数据集的子集或者 Y数据集的子集。
3. 如权利要求 1所述的方法, 其特征在于, 所述基准数据集是一个 a数据集的 子集;所述被校准数据集是一个卩数据集的子集或者一个 Y数据集的子集; 在计算校准系数 (c) 时, 以距离因子 fd做修正。
4. 如权利要求 1所述的方法, 其特征在于, 所述步骤 2) 中, 所述被校准数据 集来自于卩监测站;所述作为校准依据的基准数据集来自于被校准监测站附 近的多个移动监测站的 Y数据集;选取与被校准监测站的数据集同一时段的 基准数据子集, 合并为基准数据集。
5. 如权利要求 1所述的方法, 其特征在于, 所述步骤 2) 中, 所述被校准数据 集来自于 Y监测站;所述作为校准依据的基准数据集来自于被校准监测站附 近的多个移动监测站的 Y数据集;选取与被校准监测站的数据集同一时段的 基准数据子集, 合并为基准数据集。
6. 如权利要求 1所述的方法, 其特征在于, 所述步骤 2) 中, 先选取距离相近 的一组) g监测站和移动监测站;然后对各个) g监测站和移动监测站按可信度排 序;选取可信度最低的监测站作为被校准监测站;选取一段时间满足稳定系 数要求的数据子集作为被校准数据集;选取与被校准设备的数据集同一时段 的若干个可信度高的监测站的基准数据子集, 合并为基准数据集。
7. 如权利要求 4至 6之一所述的方法, 其特征在于, 所述步骤 3) 中基准数据 (y) 的计算, 依据基准数据集中各个数据的可信度因子和距离因子 (fd)。
8. 如权利要求 1至 6之一所述的方法, 其特征在于, 所述的稳定系数 (1) 的取 值范围为至少 80% ;所述的稳定系数 (A) 的计算公式为 : l T时间范围内落入设定区间内的基准站数据的数量 A _ T时间范围内基准站数据的数量
9. 如权利要求 1至 6之一所述的方法, 其特征在于, 所述稳定系数 (1) 与设定 区间为设定 T时间范围内的基准数据的方差有关;所述的稳定系数(1) 是否 满足要求按下面方法判断:
如果设定 T时间范围内的基准数据方差 <方差设定值 B, 则满足要求; 否则, 不满足要求。
10.如权利要求 1至 6之一所述的方法, 其特征在于, 所述稳定系数 (1) 与设定 区间为设定 T时间范围内的基准数据的标准差有关;所述的稳定系数(1) 是 否满足要求按下面方法判断:
如果设定 T时间范围内的基准数据标准差 <方差设定值 C, 则满足要求; 否则, 不满足要求。
11.如权利要求 7所述的方法, 其特征在于, 所述距离因子 (fd) 的计算方法为 :
( K
( — ' d > A
/d⑷ = ] ( 1 d , 0 < d < A 其中参数 d表示该区域几何中心点到该区域内各个站点之间的距离或该特定 位置到各个监测点的距离;该距离的设定值为 A ;在设定距离 A以内, 距离 因子为 1 ;超过设定距离 A后, 距离越远则该监测站点数据所占据的权重越 小, 距离越近则该监测站点数据所占据的权重越大;参数 K是距离权重参数, —般情况下 K = A。
12.如权利要求 1至 6之一所述的方法, 其特征在于, 所述的对比系数 (n) 的计 算公式为 :
Figure imgf000016_0001
其中, X为被校准数据, y为基准数据, r|为对比系数。
13.如权利要求 7所述的方法, 其特征在于, 所述基准数据应用归一化法进行修 正, 归一化计算公式为 :
Figure imgf000017_0001
其中 为经过修正后的基准数据;y_为未经修正的基准站数据; n为达到标准 的基准站个数。
14.如权利要求 13所述的方法, 其特征在于, 当 n=l时, 归一化计算公式为 :
yc = y - fd x (.x - y)
15.如权利要求 4至 6之一所述的方法, 其特征在于, 所述移动监测站包含主控 模块和检测模块, 所述检测模块包含至少两个同类子传感器单元组成传感器 模组;所述子传感器单元工作在正常的工作频率;所述检测模块还包含至少 一个与传感器模组同类的子传感器单元组成的低频校准模组;低频校准模组 内的子传感器单元的工作频率远低于传感器模组内子传感器单元的工作频率。
16.如权利要求 15所述的方法, 其特征在于, 所述传感器模组的工作频率与低频 校准模组的工作频率的比率为 :2:1 , 3:1 , 4:1 , 5:1 , 6:1 , 7:1 , 8:1 , 9:1 , 10:1 , 15:1 , 或者 20:1。
17.如权利要求 15所述的方法, 其特征在于, 当所述主控模块发现所述传感器模 组中一个子传感器单元出现疑似异常, 并判断该疑似异常子传感器为异常子 传感器后, 对所述异常子传感器进行隔离, 所述异常子传感器归入隔离区, 多核传感器模组降级后继续正常工作;进入隔离区的子传感器如无法自愈则 停止工作;如可以自愈则做降频工作处理, 但是子传感器输出的数据不参与 主控模块输出数据的计算;主控模块监测进入隔离区的子传感器输出的数据, 判断其是否达到恢复条件;将达到恢复条件的子传感器调离隔离区, 恢复工 作。
PCT/IB2019/051244 2018-07-25 2019-02-15 环境传感器协同校准方法 WO2020021344A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
CN201980006118.6A CN112567241A (zh) 2018-07-25 2019-02-15 环境传感器协同校准方法
PCT/CN2019/102420 WO2020043031A1 (zh) 2018-08-25 2019-08-25 一种大气污染监测传感器校准和协同工作的方法
PCT/CN2019/102419 WO2020043030A1 (zh) 2018-08-25 2019-08-25 大气污染监测设备数据可信度评价及校准方法
CN201980089854.2A CN113330283B (zh) 2018-08-25 2019-08-25 大气污染检测设备数据可信度评价及校准方法
CN201980089835.XA CN113728220B (zh) 2018-08-25 2019-08-25 一种大气污染监测传感器校准和协同工作的方法

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
IBPCT/IB2018/055531 2018-07-25
PCT/IB2018/055531 WO2019150182A1 (zh) 2018-02-01 2018-07-25 多核传感器***及其隔离和恢复的方法
CN2018109762460 2018-08-25
CN201810976246 2018-08-25

Publications (1)

Publication Number Publication Date
WO2020021344A1 true WO2020021344A1 (zh) 2020-01-30

Family

ID=69181369

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2019/051244 WO2020021344A1 (zh) 2018-07-25 2019-02-15 环境传感器协同校准方法

Country Status (2)

Country Link
CN (1) CN112567241A (zh)
WO (1) WO2020021344A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113867281A (zh) * 2021-09-09 2021-12-31 蔡丹迎 一种基于大数据的智能厂房管理方法及***
CN114323104A (zh) * 2021-12-23 2022-04-12 沈阳环境科学研究院 一种车辆环境检测传感器在线校准***

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1113268A1 (en) * 1999-12-31 2001-07-04 Moretti &amp; Carpita S.n.c. Method and apparatus for air quality monitoring in a predetermined geographical area
KR20140058727A (ko) * 2012-11-05 2014-05-15 안산시 환경 모니터링 시스템
CN105181898A (zh) * 2015-09-07 2015-12-23 李岩 基于大密度部署传感器的大气污染监控及管理方法及***
CN205483924U (zh) * 2015-09-25 2016-08-17 原点生活(北京)科技有限公司 用于检测空气中细颗粒物浓度的***
CN106644862A (zh) * 2016-09-12 2017-05-10 济南诺方电子技术有限公司 一种传感器、基于该传感器的监测站及监测站的监测方法
CN107063955A (zh) * 2017-04-18 2017-08-18 击风科技(北京)有限公司 空气颗粒物检测仪校准方法及管理***
CN206756770U (zh) * 2016-10-28 2017-12-15 神达电脑股份有限公司 空气品质监控***
CN108195728A (zh) * 2018-02-01 2018-06-22 山东诺方电子科技有限公司 一种基于多核颗粒物传感器技术的控制***及其控制方法

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5659125A (en) * 1995-06-07 1997-08-19 Nighthawk Systems, Inc. Automatic calibration method for carbon monoxide monitors
CN100575877C (zh) * 2007-12-12 2009-12-30 南京航空航天大学 基于多信息融合的航天器组合导航方法
CN101266273B (zh) * 2008-05-12 2010-11-24 徐立军 一种多传感器***故障自诊断方法
CA2828854A1 (en) * 2011-03-03 2012-09-07 Eaton Corporation Fault detection, isolation and reconfiguration systems and methods for controlling electrohydraulic systems used in construction equipment
CN102538859A (zh) * 2011-05-19 2012-07-04 广东迅通科技股份有限公司 多类传感器的监测处理方法
CN103604914B (zh) * 2013-11-11 2015-10-14 中国矿业大学 矿山物联网煤矿瓦斯移动检测数据处理方法
CN203705667U (zh) * 2013-12-26 2014-07-09 华南农业大学 无人值守基站的环境质量和气象参数自动监测装置
CN106055770B (zh) * 2016-05-26 2019-02-26 南京航空航天大学 一种基于滑模理论的航空发动机气路故障诊断方法
CN106383074A (zh) * 2016-08-31 2017-02-08 广东美的制冷设备有限公司 校准粉尘传感器的方法及设备
CN106950334B (zh) * 2017-03-06 2019-10-18 同济大学 一种空间分布式城市空气环境质量的移动监测***
CN107544459B (zh) * 2017-09-05 2020-02-14 北京控制工程研究所 一种控制***的多重故障诊断优化方法
CN107612999A (zh) * 2017-09-20 2018-01-19 广东先河科迪隆科技有限公司 大气网格化精准监控***

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1113268A1 (en) * 1999-12-31 2001-07-04 Moretti &amp; Carpita S.n.c. Method and apparatus for air quality monitoring in a predetermined geographical area
KR20140058727A (ko) * 2012-11-05 2014-05-15 안산시 환경 모니터링 시스템
CN105181898A (zh) * 2015-09-07 2015-12-23 李岩 基于大密度部署传感器的大气污染监控及管理方法及***
CN205483924U (zh) * 2015-09-25 2016-08-17 原点生活(北京)科技有限公司 用于检测空气中细颗粒物浓度的***
CN106644862A (zh) * 2016-09-12 2017-05-10 济南诺方电子技术有限公司 一种传感器、基于该传感器的监测站及监测站的监测方法
CN206756770U (zh) * 2016-10-28 2017-12-15 神达电脑股份有限公司 空气品质监控***
CN107063955A (zh) * 2017-04-18 2017-08-18 击风科技(北京)有限公司 空气颗粒物检测仪校准方法及管理***
CN108195728A (zh) * 2018-02-01 2018-06-22 山东诺方电子科技有限公司 一种基于多核颗粒物传感器技术的控制***及其控制方法

Also Published As

Publication number Publication date
CN112567241A (zh) 2021-03-26

Similar Documents

Publication Publication Date Title
WO2020043031A1 (zh) 一种大气污染监测传感器校准和协同工作的方法
CN111788578B (zh) 一种环境监测站数据可信度评价及校准方法
CN109558295B (zh) 一种性能指标异常检测方法及装置
CN108593990B (zh) 一种基于电能用户用电行为模式的窃电检测方法和应用
CN112381476B (zh) 用于确定状态异常的电能表的方法及装置
CN110383011A (zh) 出租车顶灯内多核传感器***
CN114818391A (zh) 一种基于多溯源模型的污染源浓度智能分析方法
CN110197020B (zh) 一种环境变化对水文干旱影响的分析方法
CN110927654B (zh) 一种智能电能表批次运行状态评价方法
CN106958922A (zh) 一种检测新风机滤网寿命的方法及***
CN109190967B (zh) 一种用于火力发电机组的碳排放量核算方法和***
CN107063955B (zh) 空气颗粒物检测仪校准方法及管理***
WO2020021344A1 (zh) 环境传感器协同校准方法
CN109636110A (zh) 一种获取继电保护装置运行状态的方法及装置
CN112304831B (zh) 基于质量守恒的学生行为对教室内pm2.5浓度的计算方法
CN103900666A (zh) 膜式燃气表长期运行的计量精度控制方法
CN111967717A (zh) 一种基于信息熵值的数据质量评价方法
CN105809249A (zh) 一种基于双神经网络的pm2.5浓度检测与预测***及方法
CN116739619A (zh) 一种能源电力碳排放监测分析建模方法及装置
Liu et al. Research on data correction method of micro air quality detector based on combination of partial least squares and random forest regression
CN113157684B (zh) 一种水利海量数据的查错方法
CN107145995A (zh) 生产环境安全性预测方法、装置和***
CN115423383B (zh) 基于人工智能的分布式村镇饮用水监测调控***及方法
CN107884319B (zh) 用于对粉尘传感器进行标定的***和方法
CN105488302A (zh) 一种公共建筑节能改造节能量测量方法

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE