Detection method applying Kalman algorithm

Abstract

The invention relates to a detection method applying a Kalman algorithm. The detection method comprises the following steps: acquiring environmental data; performing data iteration on the environment data through a Kalman algorithm; and uploading the iterated environment data to a cloud platform, and judging whether an alarm needs to be given or not by the cloud platform according to a preset threshold value. The method has the beneficial effects that the Kalman algorithm is utilized to predict and iterate the detected data in real time to obtain data closer to reality, so that the reliability of the data is improved, and a guarantee is provided for safe production; moreover, data are uploaded to the cloud end in real time, remote monitoring is achieved, data real-time performance and remote control capability are enhanced, monitoring efficiency and response speed are improved, and remote working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental detection, and more specifically, to a detection method using a Kalman algorithm. Background Art

[0002] In the field of modern environmental monitoring, accurate and real-time acquisition of environmental data is of great significance for protecting public health, preventing environmental pollution and early warning of disasters. Traditional environmental data detection methods mostly rely on single sensor measurements and directly read data for analysis. This method has limitations in data accuracy and real-time performance. Especially under complex and changing environmental conditions, single sensor data is easily affected by factors such as noise and interference, resulting in large data fluctuations and insufficient reliability.

[0003] In order to improve the accuracy and stability of environmental data detection, researchers have begun to explore the application of advanced mathematical algorithms to process and analyze data. Among them, the Kalman algorithm, as an efficient data fusion and prediction algorithm, has demonstrated excellent performance in many fields such as signal processing, navigation positioning and control systems. By combining the system dynamic model and observation data, the Kalman algorithm can achieve the optimal estimation of the system state, effectively suppress noise interference, and improve data accuracy.

[0004] However, there are still some challenges in applying the Kalman algorithm to the field of environmental monitoring. First, environmental monitoring involves a variety of data types, including but not limited to gas composition, noise, dust, etc. The characteristics and change patterns of these data are different, and it is necessary to establish a suitable system model in a targeted manner. Secondly, environmental data often have spatiotemporal distribution characteristics. How to effectively integrate location information (i.e., positioning data) to achieve more accurate environmental monitoring is another problem that needs to be solved. In addition, with the development of Internet of Things technology, how to efficiently upload processed environmental data to the cloud platform to achieve remote monitoring and intelligent analysis of data is also an important direction for the development of current environmental monitoring technology. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the prior art and to propose a detection method using a Kalman algorithm.

[0006] In a first aspect, a detection method using a Kalman algorithm is provided, comprising: Step 1: Obtain environmental data; Step 2: performing data iteration on the environmental data by using the Kalman algorithm; Step 3: Upload the iterated environmental data to the cloud platform, which determines whether an alarm is needed based on the preset threshold.

[0007] Preferably, in step 1, the environmental data is any one of gas composition data, noise data or dust data.

[0008] Preferably, in step 1, the environmental data also includes positioning data.

[0009] Preferably, step 2 comprises: Step 2.1, establishing an environment detection system model, wherein the environment detection system model includes an environment detection state equation and an environment detection observation equation; Step 2.2, initializing system parameters, wherein the system parameters include state estimation and state covariance equation; Step 2.3, based on the state estimation value and state covariance equation at the previous moment, predict the state estimation value and state covariance equation at the next moment; Step 2.4, obtain the detection target observation value at the current moment, and calculate the observation residual; obtain the covariance matrix of the observation noise, and use the Kalman gain to update the state estimate and the state covariance matrix based on the observation residual and the covariance matrix of the observation noise; Step 2.5: Repeat steps 2.3 to 2.4 to obtain the final state estimate after reaching the preset number of iterations.

[0010] Preferably, in step 3, the cloud platform displays the iterated environmental data in the form of a line graph.

[0011] Preferably, in step 3, the cloud platform generates an alarm window when an alarm is needed, and the alarm information in the alarm window includes: alarm time and environmental data exceeding a preset threshold.

[0012] In a second aspect, a detection device using a Kalman algorithm is provided, which is used to execute any of the methods described in the first aspect, including: a detection module, a base, a data processing module, a control button, and a display screen; The detection module is installed on the base, and the base has the data processing module and the battery built in; the control buttons and the display screen are arranged on the top surface of the base.

[0013] In a third aspect, a detection system using a Kalman algorithm is provided, which is used to execute any of the methods described in the first aspect, including: Acquisition module, used to obtain environmental data; An iteration module, used for performing data iteration on the environmental data by using a Kalman algorithm; The upload module is used to upload the iterated environmental data to the cloud platform. The cloud platform determines whether an alarm is needed based on the preset threshold.

[0014] According to a fourth aspect, a computer storage medium is provided, wherein a computer program is stored in the computer storage medium; when the computer program is executed on a computer, the computer executes any method described in the first aspect.

[0015] The beneficial effects of the present invention are: 1. The present invention uses the Kalman algorithm to predict and iterate the detected data in real time to obtain data that is closer to the actual data, thereby improving the reliability of the data and ensuring safe production.

[0016] 2. The modular design of the present invention is easy to disassemble and replace, and enhances portability and flexibility. For different detection objects, just replace the module, which simplifies the operation, reduces maintenance costs, and is efficient and practical.

[0017] 3. The present invention uploads data to the cloud in real time to achieve remote monitoring, enhance data real-time and remote control capabilities, improve monitoring efficiency and response speed, and improve the efficiency of remote work. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of an environment detection system provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of an environment detection device provided by an embodiment of the present invention; Figure 3 A flow chart of a detection method using a Kalman algorithm provided in an embodiment of the present invention; Figure 4 A Kalman filter algorithm workflow diagram provided by an embodiment of the present invention; Figure 5 A functional schematic diagram of a cloud platform provided by an embodiment of the present invention; Description of reference numerals: detection module 1 , base 2 , data processing module 3 , control button 4 , display screen 5 . DETAILED DESCRIPTION

[0019] The present invention is further described below in conjunction with embodiments. The description of the following embodiments is only used to help understand the present invention. It should be noted that for ordinary persons in the art, without departing from the principle of the present invention, the present invention can also be modified in some ways, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

[0020] Embodiment 1: Embodiment 1 of the present application provides a detection device using the Kalman algorithm, such as Figure 2 As shown, it includes: a detection module 1, a base 2, a data processing module 3, a control button 4 and a display screen 5; The detection module 1 is installed on the base 2 , and may be a gas detection module, a noise detection module, etc. The base 2 has the data processing module 3 and a battery built in; the control button 4 and the display screen 5 are arranged on the top surface of the base 2 .

[0021] It should be noted that the detection device can be set on the brim of the safety helmet to perform environmental detection.

[0022] In addition, the embodiment of the present application also provides an environmental detection system, including a cloud platform and the above-mentioned detection device. Specifically, the operation process of the detection system can be divided into three parts: environmental detection, Kalman filter screening and cloud monitoring. Figure 1 As shown, the detection system can be divided into physical layer, detection layer, data layer, and cloud application layer.

[0023] Among them, Figure 3 As shown in the figure, the physical layer represents the actual work scene and physical objects, such as the on-site air and noise environment, gas detection module, noise detection module, safety helmet, etc. Physical entities such as the gas and noise detection chips in the gas detection module and the noise detection module perform analog-to-digital conversion on the gas components and environmental noise in the environment, and perform GPS positioning on themselves through the antenna, and summarize some task conditions such as gas composition data, noise data, dust data, positioning data, etc. in the detection layer and pass them to the data layer.

[0024] The detection layer represents detection data, that is, data collected through chip sensors, including the oxygen content in the ambient air, the content of toxic and harmful gases such as carbon monoxide, dust concentration, ambient noise, GPS positioning information, etc. These data form the basis for the display of the environmental detection system in the cloud.

[0025] The data layer mainly uses the Kalman filter algorithm to iteratively analyze and process the data. Specifically, it includes functional modules such as data prediction, system correction, and data iteration. Taking gas detection as an example, the Kalman filter algorithm can be used to process the raw data output by the sensor and extract more accurate gas concentration information. For example, when detecting toxic or combustible gases, the Kalman filter algorithm can reduce the impact of sensor noise and drift and improve the reliability and accuracy of detection. Similarly, similar effects can be achieved in applications such as noise and dust detection.

[0026] The cloud application layer is the terminal layer, which is mainly used to realize display, monitoring and other functions. Figure 5 As shown, the cloud application layer specifically includes functions such as data line chart display and operation and maintenance monitoring.

[0027] Embodiment 2: On the basis of Example 1, Example 2 of the present application provides a detection method using the Kalman algorithm, including: Step 1: Get environmental data.

[0028] In step 1, the environmental data is any one of gas composition data, noise data or dust data. Specifically, the environmental data may include the concentration of gases such as oxygen, carbon monoxide, sulfur dioxide, and nitric oxide in the current gas environment; the concentration of dust in the gas environment; the state of gas concentration change: the gas concentration increases or decreases over time in different environments; and the data of noise in the current environment: noise frequency, noise sound pressure, noise power and other data; the state of noise change: the noise increases or decreases over time in different environments.

[0029] In addition, in step 1, the environmental data also includes GPS positioning data, that is, the geographical location information of the current chip, which is presented in the form of longitude and latitude, which helps to accurately locate the position of the staff.

[0030] Step 2: Perform data iteration on the environmental data using the Kalman algorithm.

[0031] like Figure 4 As shown, step 2 includes: Step 2.1: Establish an environment detection system model, wherein the environment detection system model includes an environment detection state equation and an environment detection observation equation.

[0032] The state equation describes the transition of the system state (observation of the detection target) from one time step to the next. This equation may include the dynamic characteristics of the system, such as the natural attenuation of the observation value, the influence of external sources (such as gas leakage or noise interference), and possible disturbances (such as environmental changes). The observation equation describes the relationship between the sensor output (observation value) and the system state.

[0033] The formula of the state equation is:

[0034] in, is the system state at time k (such as gas concentration or noise frequency). is the state transition matrix, describing how the state changes from The time is transferred to the k moment. is the control input matrix, is the control input (such as a change in gas source or a change in noise source). is the process noise, and its covariance is .

[0035] The formula of the observation equation is:

[0036] in, yesk An observation at a moment in time (such as a sensor reading). is the observation matrix, describing how the system states are mapped to observations. is the observation noise, and its covariance is .

[0037] Step 2.2: Initialize system parameters, which include state estimation and state covariance equation.

[0038] Before starting the filtering process, it is necessary to predict the system's state estimate (such as the initial observation estimate) and the state covariance matrix (representing the uncertainty of the state estimate). These initialization values ​​are usually based on prior knowledge or the initial conditions when the system starts. For example, the initial state estimate is expressed as , the initial state covariance matrix is ​​expressed as .

[0039] Step 2.3: Based on the state estimate and state covariance equation at the previous moment, predict the state estimate and state covariance equation at the next moment.

[0040] Specifically, at each time step, the Kalman filter first predicts the system state at the current time step based on the system state equation and the state estimate of the previous time step. At the same time, the gas concentration state at the next moment is also predicted based on the system's dynamic model (such as gas diffusion model, noise propagation model, sensor response model, etc.). This usually involves multiplying the current state estimate with the system's dynamic equation (such as the state transfer matrix) and may include external control inputs (such as changes in gas sources or noise sources).

[0041] The formula for predicting the status is:

[0042] in, is the predicted state at time k (based on the information at time k-1). It is the state transition matrix, which describes how the system state is transferred from the previous moment to the current moment. yes The optimal state estimate at time. is the control input matrix, yes Control input at any time.

[0043] The formula for the forecast covariance is:

[0044] in, is the forecast covariance matrix at time k. yes The optimal covariance matrix at time . is the covariance matrix of the process noise.

[0045] Step 2.4, obtain the detection target observation value at the current moment, and calculate the observation residual (i.e., the difference between the observation value and the predicted value); obtain the covariance matrix of the observation noise, and based on the observation residual and the covariance matrix of the observation noise, use the Kalman gain to update the state estimate and the state covariance matrix.

[0046] Among them, Kalman gain is the proportional relationship between prediction error and measurement residual, which determines the weight of prediction value and observation value when updating state estimation. Covariance update is used to reflect the uncertainty of updated state estimation.

[0047] The formula for calculating the Kalman gain is:

[0048] in, is the Kalman gain at time k. is the observation matrix, which describes how the system state is mapped to the observation space. is the covariance matrix of the observation noise.

[0049] The formula for updating the state estimate is:

[0050] The formula for updating the state covariance matrix is:

[0051] in, is the optimal covariance matrix at time k. is the identity matrix.

[0052] Step 2.5: Repeat steps 2.3 to 2.4 to obtain the final state estimate after reaching the preset number of iterations.

[0053] The measurement and observation update steps are iterated continuously, and the state estimate is updated according to the new observation value at each time step. As the iterations proceed, the error of the state estimate will gradually decrease, resulting in a more accurate estimate of the detected target.

[0054] Step 3: Upload the iterated environment data to the cloud platform, such as Figure 3 As shown, the cloud platform determines whether an alarm is needed based on a preset threshold.

[0055] The basic device information module of the cloud platform will display device details, such as device power, SIM card ICCID and other basic information. A device sensor snapshot function is provided to instantly present the latest data collected by the sensor to ensure that users obtain the latest status information. It also has GPS positioning information, which can provide the latest location of the device on the map and the original coordinate value of the WGS_84 coordinate system.

[0056] The cloud has the ability to record the collection history and can display the device's most recent 500 collection histories, including sensor values ​​and device status parameters.

[0057] The cloud platform supports trend chart viewing of device sensors. Users can switch months as needed to view trend changes in different time periods. To optimize rendering efficiency, the cloud automatically performs 10-minute average sampling for charts with a time span of more than 24 hours to ensure smooth loading of charts while retaining key data features. Through the chart, users can intuitively and quickly understand the information collected by the sensor, and because the Kalman filter algorithm is applied, the chart information will be more accurate.

[0058] In addition, the cloud platform integrates advanced real-time alarm functions to ensure rapid response when key collected information exceeds the preset threshold range. Users can set the safety threshold range for each monitored target on the cloud platform. These thresholds are flexibly adjusted according to the needs of the actual application scenario to ensure the accuracy and effectiveness of monitoring. The system continuously monitors the collected information uploaded by the device and compares it with the preset threshold range in real time. Once it is found that the observation value of a certain detection target exceeds the safety range, the system will immediately trigger the alarm mechanism. When the observation value of the detection target exceeds the threshold, a striking alarm window will pop up automatically. The window contains detailed alarm information, such as the type of gas exceeding the threshold, the current concentration value, the alarm time, etc., so that users can quickly understand and deal with the alarm. When the alarm is triggered, in addition to the pop-up window prompt of the cloud platform, sound or light signals can also be used to remind the on-site personnel to pay attention. Combined with the alarm function, the system can further integrate the emergency response process. Once the alarm is triggered, the system automatically executes the preset emergency measures, such as sending alarm SMS or emails to relevant personnel, starting backup equipment, etc., to minimize potential risks.

[0059] In the embodiment of the present application, a background monitoring center can also be set up, and according to the monitoring needs, the key data can be summarized and calculated in the form of charts and displayed in a unified panel to meet the needs of centralized monitoring.

[0060] It should be noted that the parts in this embodiment that are the same or similar to those in Embodiment 1 can be referenced to each other and will not be described in detail in this application.

[0061] Embodiment 3: Based on Example 2, Example 3 of the present application provides a detection system using the Kalman algorithm, including: Acquisition module, used to obtain environmental data; An iteration module, used for performing data iteration on the environmental data by using a Kalman algorithm; The upload module is used to upload the iterated environmental data to the cloud platform. The cloud platform determines whether an alarm is needed based on the preset threshold.

[0062] Specifically, the system provided in this embodiment is a system corresponding to the method provided in Example 2. Therefore, the parts in this embodiment that are the same or similar to those in Example 2 can be referenced to each other and will not be repeated in this application.

Claims

1. A detection method using Kalman algorithm, characterized in that: include: Step 1: Obtain environmental data; Step 2: performing data iteration on the environmental data by using the Kalman algorithm; Step 3: Upload the iterated environmental data to the cloud platform, which determines whether an alarm is needed based on the preset threshold.

2. The detection method using the Kalman algorithm according to claim 1, characterized in that: In step 1, the environmental data is any one of gas composition data, noise data or dust data.

3. The detection method using the Kalman algorithm according to claim 2, characterized in that: In step 1, the environmental data also includes positioning data.

4. The detection method using the Kalman algorithm according to claim 3, characterized in that: Step 2 includes: Step 2.1, establishing an environment detection system model, wherein the environment detection system model includes an environment detection state equation and an environment detection observation equation; Step 2.2, initializing system parameters, wherein the system parameters include state estimation and state covariance equation; Step 2.3, based on the state estimation value and state covariance equation at the previous moment, predict the state estimation value and state covariance equation at the next moment; Step 2.4, obtain the detection target observation value at the current moment, and calculate the observation residual; obtain the covariance matrix of the observation noise, and use the Kalman gain to update the state estimate and the state covariance matrix based on the observation residual and the covariance matrix of the observation noise; Step 2.5: Repeat steps 2.3 to 2.4 to obtain the final state estimate after reaching the preset number of iterations.

5. The detection method using the Kalman algorithm according to claim 4, characterized in that: In step 3, the cloud platform displays the iterated environmental data in the form of a line graph.

6. The detection method using the Kalman algorithm according to claim 4, characterized in that: In step 3, the cloud platform generates an alarm window when an alarm is needed, and the alarm information in the alarm window includes: alarm time and environmental data exceeding a preset threshold.

7. A detection device using Kalman algorithm, characterized in that: Used to execute the method according to any one of claims 1 to 6, comprising: a detection module (1), a base (2), a data processing module (3), a control button (4) and a display screen (5); The detection module (1) is mounted on the base (2), and the base (2) has the data processing module (3) and a battery built in; the control button (4) and the display screen (5) are arranged on the top surface of the base (2).

8. A detection system using a Kalman algorithm, characterized in that: Used to perform the method according to any one of claims 1 to 6, comprising: Acquisition module, used to obtain environmental data; An iteration module, used for performing data iteration on the environmental data by using a Kalman algorithm; The upload module is used to upload the iterated environmental data to the cloud platform. The cloud platform determines whether an alarm is needed based on the preset threshold.

9. A computer storage medium, characterized in that The computer storage medium stores a computer program; when the computer program is executed on a computer, the computer executes any one of the methods described in claims 1 to 6.

Images