Internet of Things edge control system applied to sewage treatment field

Abstract

The invention provides an Internet of Things edge control system applied to the field of sewage treatment, and relates to the technical field of data management. According to the invention, an HMI scheme is realized through cooperation of the mobile terminal and the intelligent gateway, field touch screen equipment is replaced, the hardware cost is reduced, and the operation convenience and the application richness are improved. According to the system, software is adopted to replace a hardware touch screen controller, reliability and safety are improved, meanwhile, operation logs are recorded in the mobile handheld device and the intelligent gateway, backtracking inspection is facilitated, and interaction support is provided for follow-up parameter configuration and device initialization; in the aspect of parameter configuration, the system introduces an operation effect as key information, and is combined with daily work of configuring operation parameters by experts, so that the efficiency of configuring control parameters is improved, and an effective expert experience library is formed; in the aspect of equipment initialization, the system performs filter parameter initialization by using cloud historical data, so that the filter initialization time is saved, and the prediction accuracy and the response speed are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to an Internet of Things edge control system applied to the field of sewage treatment. Background Art

[0002] With the rapid development of industrial automation and intelligence, edge computing equipment, as an important equipment in the field of sewage treatment, has become an inevitable trend in the development of the industry through technological upgrading and digital transformation. However, the sewage treatment field is still facing many challenges and limitations in terms of IoT edge control.

[0003] Before computing, edge computing devices need to go through the three steps of "device control-parameter configuration-device initialization". However, due to the complexity of sewage treatment scenarios, certain problems will be encountered during these three steps; such as:

[0004] When controlling the equipment, we need to use a field controller (such as a touch screen, etc.) for on-site control. However, the sewage treatment scenario is complex, and it is difficult for us to conveniently operate some devices that are installed in remote or dangerous locations.

[0005] When configuring parameters, the software, algorithms, and parameters of the current IoT + edge computing solutions are deeply coupled. Industry experts are not good at designing software and algorithms, but are only good at setting parameters such as analog quantities and digital quantities related to the specific operating environment and control objectives of professional scenarios, which makes it very difficult for industry experts to get started.

[0006] When the equipment is initialized, the sewage treatment field has a time-sensitive demand for processing data; however, under the existing edge computing architecture, when the equipment is initialized, the initial value of the filter is usually selected using true value, expectation, and random value, but the true value is difficult to obtain, and the expectation is suitable for biased estimation. The convergence speed will be affected by using random values. If the data collection cycle is at the minute level, the filter often takes more than 10 minutes to converge.

[0007] If these problems are not solved, it will be difficult for the Internet of Things technology to be implemented in the field of sewage treatment; therefore, it is necessary to provide an Internet of Things edge control system applied to the field of sewage treatment to solve the above technical problems. Summary of the invention

[0008] In order to overcome the shortcomings of the existing technology, the present invention provides an Internet of Things edge control system applied to the field of sewage treatment, which mainly solves the problems encountered by the existing technology in the three links of "device control-parameter configuration-device initialization" due to the complexity of sewage treatment scenarios.

[0009] The purpose of the present invention is achieved through the following technical solutions:

[0010] Provided is an Internet of Things edge control system applied to the field of sewage treatment, wherein an intelligent gateway is set in the sewage treatment scene, the intelligent gateway is used to configure one or more wireless local area networks, the wireless local area network is used to establish an intranet communication connection with the corresponding edge computing device in the sewage treatment scene, and to establish an extranet communication connection with the cloud through the intelligent gateway, the cloud is used to provide cloud services; wherein,

[0011] The smart gateway is provided with a local HMI server, and a mobile handheld device connection subnet is provided in the wireless local area network. The user installs an HMI client on the mobile handheld device, and the HMI client supports remote HMI mode and local HMI mode;

[0012] When performing local device control: the user accesses the mobile handheld device connection subnet through the HMI client, sends a request to the intelligent gateway, obtains the edge computing device under its jurisdiction, enters the local HMI mode, and controls the edge computing device under its jurisdiction on the mobile handheld device;

[0013] The cloud is provided with a control parameter database and a priority recommendation module, and the HMI client is provided with an effect management module and a control parameter design module; wherein the user sets the effect parameter through the effect management module and sets the control parameter through the control parameter design module; the cloud saves the control parameter configuration through the control parameter database, and recommends the control parameter according to the effect parameter through the priority recommendation module;

[0014] A historical time series database is set on the cloud, and an edge controller is set in the edge computing device; wherein, the edge controller is used to feed back initial data to the cloud and obtain filter parameters, and the historical time series database is used to store and query filter parameters, and device initialization is carried out through the cooperation between the edge controller and the historical time series database.

[0015] As a further solution, the HMI client controls the device through the following steps:

[0016] The user enters the industrial control area and starts the local control mode through the HMI client;

[0017] The HMI client connects to the mobile handheld device connection subnet and sends an authentication request to the cloud;

[0018] The HMI client asks the cloud whether the authentication is successful;

[0019] If so, obtain on-site control authority for the edge computing device under its jurisdiction and enter local control mode;

[0020] If not, exit the local control mode;

[0021] Local control: The HMI instructions corresponding to the device control are sent to the smart gateway HMI server, and the smart gateway executes them on behalf of the device, recording and uploading the operation logs throughout the process until the local control mode is exited.

[0022] As a further solution, the local HMI server performs device control through the following steps:

[0023] When an HMI client is connected to the mobile handheld device connection subnet, the authentication procedure is started;

[0024] Send a check to the HMI client to check whether it is passed;

[0025] When authentication is passed, the local control permissions of the edge computing devices under the jurisdiction are opened to the HMI client;

[0026] Receive HMI commands sent by the HMI client, convert the HMI commands to the field bus of the corresponding edge computing device, and record and upload the operation log throughout the process.

[0027] As a further solution, the cloud is also provided with an effect value database and a data sample library; wherein, the effect value database is used to store the effect values ​​corresponding to the control parameter configurations, and the data sample library is used to save the matching results of the queried optimal effect values ​​and the control parameter configurations, and the control parameter configurations have a unique matching identifier in each database.

[0028] As a further solution, perform parameter configuration as follows:

[0029] The user starts the HMI client and connects to the cloud after authentication;

[0030] Input the effect parameters required by the user through the effect management module and feed them back to the cloud;

[0031] The cloud uses the priority recommendation module to query the effect value that meets the matching threshold from the effect value database using the optimal matching algorithm based on the effect parameters required by the user;

[0032] Obtain one or more control parameter configurations through the association relationship of the effect values ​​and feed them back to the HMI client;

[0033] The user selects the optimal effect value through the HMI client and refreshes the matching results stored in the data sample library;

[0034] The HMI client asks the user whether to directly use the control parameter configuration corresponding to the optimal effect value;

[0035] If yes, the control parameter configuration is directly issued and an execution record is saved in the control parameter database;

[0036] If not, the user modifies the control parameter configuration through the HMI client and saves an execution record in the control parameter database;

[0037] The HMI client generates HMI instructions through control parameter configuration, and converts the HMI instructions to the field bus of the corresponding edge computing device through the local HMI server of the smart gateway;

[0038] The edge computing device runs the control parameter configuration and collects the running effect, uploads the effect value to the effect value database, and saves an execution record;

[0039] The cloud saves the matching relationship between the control parameter configuration and the effect value in the execution record.

[0040] As a further solution, perform device initialization as follows:

[0041] Start the edge computing device and connect it to the cloud through the smart gateway relay to obtain cloud services;

[0042] The edge controller reads the initial sequence of sensor measurements, uploads them to the cloud as initial data, and continues to measure data;

[0043] The cloud uses the priority recommendation module to query the historical measurement value sequence that meets the matching threshold from the historical time series database based on the initial measurement value sequence using the optimal matching algorithm;

[0044] Setting a cloud Kalman filter in the cloud and inputting a historical measurement value sequence into the cloud Kalman filter;

[0045] Initialize the cloud Kalman filter in the cloud and record the initialized filter parameters;

[0046] The cloud sends the filter parameters to the edge computing device, and the edge controller sets the local Kalman filter and controls the operation of the local Kalman filter according to the filter parameters;

[0047] The local Kalman filter is continuously run, the corresponding filter parameters are output and uploaded to the historical time series database for storage.

[0048] As a further solution, the priority recommendation module uses a moving window average method to perform a reverse search or screens the time series by the mean / median to set the optimal matching algorithm.

[0049] As a further solution, the intelligent gateway also performs hidden settings for the wireless local area network and authenticates the edge computing device; the edge device is connected to the hidden wireless local area network through an encrypted channel.

[0050] Compared with related technologies, the IoT edge control system applied to sewage treatment provided by the present invention has the following advantages:

[0051] The present invention first replaces the on-site touch screen device through the HMI solution implemented by the mobile terminal and the intelligent gateway in a collaborative manner, reduces the hardware cost, and improves the convenience of operation and the richness of application. The system uses software to replace the hardware touch screen controller to improve reliability and safety. At the same time, the operation log is recorded in both the mobile handheld device and the intelligent gateway, which is convenient for retrospective inspection and provides interactive support for subsequent parameter configuration and equipment initialization. In terms of parameter configuration, the system introduces the operation effect as key information, combines it with the daily work of experts configuring the operation parameters, improves the efficiency of configuring the control parameters, and forms an effective expert experience library. In terms of equipment initialization, the system uses cloud historical data to initialize the filter parameters, saving the filter initialization time and improving the prediction accuracy and response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of the structure of an Internet of Things edge control system applied to the field of sewage treatment provided by the present invention;

[0053] Figure 2 A schematic diagram of the device control process provided by the present invention;

[0054] Figure 3 A flow chart of parameter configuration provided by the present invention;

[0055] Figure 4 A schematic diagram of the device initialization process provided by the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0057] See also Figure 1 The present invention provides an Internet of Things edge control system applied to the field of sewage treatment, in which an intelligent gateway is set in the sewage treatment scene, and the intelligent gateway is used to configure one or more wireless local area networks, and the wireless local area network is used to establish an intranet communication connection with the corresponding edge computing device in the sewage treatment scene, and to establish an external network communication connection with the cloud through the intelligent gateway, and the cloud is used to provide cloud services; wherein,

[0058] The smart gateway is provided with a local HMI server, and a mobile handheld device connection subnet is provided in the wireless local area network. The user installs an HMI client on the mobile handheld device, and the HMI client supports remote HMI mode and local HMI mode;

[0059] When performing local device control: the user accesses the mobile handheld device connection subnet through the HMI client, sends a request to the intelligent gateway, obtains the edge computing device under its jurisdiction, enters the local HMI mode, and controls the edge computing device under its jurisdiction on the mobile handheld device;

[0060] The cloud is provided with a control parameter database and a priority recommendation module, and the HMI client is provided with an effect management module and a control parameter design module; wherein the user sets the effect parameter through the effect management module and sets the control parameter through the control parameter design module; the cloud saves the control parameter configuration through the control parameter database, and recommends the control parameter according to the effect parameter through the priority recommendation module;

[0061] A historical time series database is set on the cloud, and an edge controller is set in the edge computing device; wherein, the edge controller is used to feed back initial data to the cloud and obtain filter parameters, and the historical time series database is used to store and query filter parameters, and device initialization is carried out through the cooperation between the edge controller and the historical time series database.

[0062] This embodiment mainly solves the problems existing in the three links of "device control-parameter configuration-device initialization" in the existing edge computing architecture in the sewage treatment scenario.

[0063] In terms of device control:

[0064] Professional technicians can remotely monitor and control sewage treatment equipment on site through remote terminals, but touch screen equipment still needs to be retained at the sewage treatment site to enable emergency control in the event of network interruption or abnormal conditions in the cloud service system, thereby improving system resilience. These touch screens are installed in a fixed manner and are connected to the control system through an industrial interface (such as RS485 / USB) or industrial Ethernet (such as Profinet). Due to the fixed position, operation is inconvenient and hardware costs are increased.

[0065] With the development of edge computing technology, the functions of edge intelligent gateways deployed on site are becoming more and more abundant. In addition to the northbound network interface and the southbound industrial control interface, wireless LAN functions such as Wifi / Bluetooth can also be easily implemented. Its reliability and security have also been greatly improved, making it possible for mobile terminal devices such as mobile phones and tablets to securely access the on-site control network.

[0066] like Figure 2As shown, the HMI solution implemented by the mobile terminal + intelligent gateway in this embodiment is used to replace the touch screen equipment on the sewage treatment site, and reduce the hardware cost of the on-site control system while ensuring safety and reliability; this embodiment consists of an intelligent gateway with a wireless gateway function, a mobile handheld device with a corresponding wireless connection function (such as a mobile phone) and an original edge computing device.

[0067] When the mobile handheld device is far away from the site, it can access the smart gateway through the Internet of Things to the north for remote cloud control. When it is on site, it can access the wireless LAN of the smart gateway to the south for local control. Among them, the smart gateway has at least three types of corresponding interfaces: a network interface that can access cloud services to the north, a wireless LAN interface that can receive secure access from mobile handheld devices to the south, and an industrial interface that can monitor and control edge computing devices.

[0068] Compared with the prior art, this embodiment uses software to replace the hardware touch screen controller, which has lower hardware cost, more convenient operation and richer applications; the wireless LAN access operation device has higher reliability than cloud access, the intelligent gateway secondary authentication improves security, the hidden LAN + username and password realizes access authentication, and the local HMI server authenticates the HMI client operation authority; the operation log can be recorded on the mobile handheld device and the intelligent gateway at the same time, which is convenient for tracing back the inspection record.

[0069] In terms of parameter configuration:

[0070] With the development of technologies such as the Internet of Things, big data, and cloud services, and with the increasing intelligence of on-site controllers, manual remote control no longer requires repetitive and boring equipment switching operations. On-site control experts only need to use remote terminals to design control algorithms for each station and send them to on-site intelligent controllers through cloud services for intelligent execution.

[0071] However, industry experts are not good at designing software and algorithms. They are only good at setting parameters such as analog quantities and digital quantities related to the specific operating environment and control objectives of professional scenarios. The corresponding software, algorithms, and parameters of the current IoT + edge computing solutions are deeply coupled, making it difficult for industry experts to get started. There are two main implementations:

[0072] The first method is to combine parameter configuration with algorithm and software, package them as (parameter + algorithm) or (parameter + algorithm + software), and manage them according to version numbers. In this case, if a parameter is modified, the software package or algorithm package must be upgraded. Once the parameter configuration is modified frequently, there will be too many versions, and a large number of repeated or similar parameter configurations will be difficult to manage.

[0073] The second method is to create a series of parameter templates, and experts select the most suitable templates. Based on this template, they modify and optimize the parameters to form different versions. In essence, this is to classify the parameters. There are still too many versions in each category, and the same repeated or similar parameter configurations are difficult to manage.

[0074] Therefore, the original software / algorithm + parameter packaging management method did not distinguish between software, algorithms, and parameters. Software corresponds to the general computer field, algorithms correspond to industry-wide models, and parameters correspond to specific industry application scenarios. The main work of industry experts is to design industry-wide algorithms and adjust parameters for specific application scenarios. Packaging parameters and algorithms / software for management leads to a huge number of software / algorithm versions, which cannot be reused quickly and can only be reconfigured. The control parameters will be repeatedly exacerbated.

[0075] For this reason, Figure 3 As shown, this embodiment introduces the operating effect as key information, which is combined with the daily work of experts configuring operating parameters without adding extra workload; in the original version management algorithm, the number of historical versions increases linearly, while this embodiment only recommends the best few according to the operating effect target each time, which can greatly improve the efficiency of experts configuring control parameters; the set effect is used as the annotation input to easily obtain training data samples, save data annotation costs, and form an effective expert experience library.

[0076] In addition, this embodiment uses the optimal recommendation algorithm to sort the huge version list according to the operating results and turn it into a few optimal or suboptimal options; it combines the data sample annotation work with the daily parameter adjustment work to generate a sample database in the daily work of experts.

[0077] In terms of device initialization:

[0078] Edge computing is a distributed computing architecture that has been formed with the development of technologies such as the Internet of Things, big data, and cloud services. It can well combine the characteristics of large data volume and sufficient computing power in the cloud with the high real-time and strong reliability of the edge.

[0079] The Kalman filter algorithm and the extended Kalman filter algorithm are usually used in the field of industrial control to filter or predict continuous analog quantities. During the operation of the algorithm, due to external noise interference and uncontrollable environmental factors, adaptive Kalman filtering is also used to estimate and correct the model and noise statistical characteristics while using the measurement data, thereby modifying the filter design and filtering error.

[0080] In the edge computing architecture, edge intelligent controllers usually undertake data collection, filtering, compression and transmission tasks. However, when the system is initialized, the initial value of the filter is usually selected using true value, expectation, and random value. However, true value is difficult to obtain, and expectation is suitable for biased estimation. The convergence speed will be affected by using random value. If the data collection cycle is in minutes, the filter often takes more than 10 minutes to converge.

[0081] However, a large amount of historical data is stored in the cloud. If the historical data in the cloud is used as a priori data, the filter parameters are initialized in the cloud, and then the corresponding parameters and initial values ​​are sent to the edge server, it only takes a few seconds, which can greatly save the filter initialization time.

[0082] Therefore, the Kalman filter algorithm running on the existing edge controller is difficult to achieve fast convergence during the startup initialization phase due to the lack of true value; if fast convergence is to be achieved, the filter parameters must strengthen the trust in the measured values, but this will inevitably affect the filtering effect.

[0083] like Figure 4 As shown, the present invention, with the help of edge computing technology, uses the time series characteristics (mean, median and change trend) of a small amount of measurement data to perform a moving window reverse search in the cloud historical database, and the searched data is input into the Kalman filter to finally obtain the predicted value and filter parameters close to the true value, which are sent to the edge controller to replace the initial filter value and parameters of the edge controller; therefore, the true value of the present invention comes from the prior probability estimation, and the initialization convergence stage is directly skipped with a high probability; the model parameters are also obtained by simulating the operation of historical data, which can improve the prediction accuracy and response speed at the same time.

[0084] Note: The search for historical data can use algorithms such as Euclidean distance, variance, long short-term memory, etc. The technology is completely feasible and will not be elaborated here.

[0085] The above embodiments only express preferred implementation modes, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention.

Claims

1. An Internet of Things edge control system applied in the field of sewage treatment, characterized in that: An intelligent gateway is set in the sewage treatment scene, and the intelligent gateway is used to configure one or more wireless local area networks. The wireless local area network is used to establish an intranet communication connection with the corresponding edge computing device in the sewage treatment scene, and to establish an extranet communication connection with the cloud through the intelligent gateway. The cloud is used to provide cloud services; wherein, The smart gateway is provided with a local HMI server, and a mobile handheld device connection subnet is provided in the wireless local area network. The user installs an HMI client on the mobile handheld device, and the HMI client supports remote HMI mode and local HMI mode; When performing local device control: the user accesses the mobile handheld device connection subnet through the HMI client, sends a request to the intelligent gateway, obtains the edge computing device under its jurisdiction, enters the local HMI mode, and controls the edge computing device under its jurisdiction on the mobile handheld device; The cloud is provided with a control parameter database and a priority recommendation module, and the HMI client is provided with an effect management module and a control parameter design module; wherein the user sets the effect parameter through the effect management module and sets the control parameter through the control parameter design module; the cloud saves the control parameter configuration through the control parameter database, and recommends the control parameter according to the effect parameter through the priority recommendation module; A historical time series database is set on the cloud, and an edge controller is set in the edge computing device; wherein, the edge controller is used to feed back initial data to the cloud and obtain filter parameters, and the historical time series database is used to store and query filter parameters, and device initialization is carried out through the cooperation between the edge controller and the historical time series database.

2. According to claim 1, an Internet of Things edge control system applied to the field of sewage treatment is characterized in that: The HMI client controls the device through the following steps: The user enters the industrial control area and starts the local control mode through the HMI client; The HMI client connects to the mobile handheld device connection subnet and sends an authentication request to the cloud; The HMI client asks the cloud whether the authentication is successful; If so, obtain on-site control authority for the edge computing device under its jurisdiction and enter local control mode; If not, exit the local control mode; Local control: The HMI instructions corresponding to the device control are sent to the smart gateway HMI server, and the smart gateway executes them on behalf of the device, recording and uploading the operation logs throughout the process until the local control mode is exited.

3. The IoT edge control system applied to sewage treatment according to claim 1 is characterized in that: The local HMI server controls the device through the following steps: When an HMI client is connected to the mobile handheld device connection subnet, the authentication procedure is started; Send a check to the HMI client to check whether it is passed; When authentication is passed, the local control permissions of the edge computing devices under the jurisdiction are opened to the HMI client; Receive HMI commands sent by the HMI client, convert the HMI commands to the field bus of the corresponding edge computing device, and record and upload the operation log throughout the process.

4. The IoT edge control system applied to sewage treatment according to claim 1 is characterized in that: The cloud is also provided with an effect value database and a data sample library; wherein the effect value database is used to store the effect values ​​corresponding to the control parameter configurations, and the data sample library is used to save the matching results of the queried optimal effect values ​​and the control parameter configurations, and the control parameter configurations have a unique matching identifier in each database.

5. The IoT edge control system applied to sewage treatment according to claim 4 is characterized in that: Perform parameter configuration as follows: The user starts the HMI client and connects to the cloud after authentication; Input the effect parameters required by the user through the effect management module and feed them back to the cloud; The cloud uses the priority recommendation module to query the effect value that meets the matching threshold from the effect value database using the optimal matching algorithm based on the effect parameters required by the user; Obtain one or more control parameter configurations through the association relationship of the effect values ​​and feed them back to the HMI client; The user selects the optimal effect value through the HMI client and refreshes the matching results stored in the data sample library; The HMI client asks the user whether to directly use the control parameter configuration corresponding to the optimal effect value; If yes, the control parameter configuration is directly issued and an execution record is saved in the control parameter database; If not, the user modifies the control parameter configuration through the HMI client and saves an execution record in the control parameter database; The HMI client generates HMI instructions through control parameter configuration, and converts the HMI instructions to the field bus of the corresponding edge computing device through the local HMI server of the smart gateway; The edge computing device runs the control parameter configuration and collects the running effect, uploads the effect value to the effect value database, and saves an execution record; The cloud saves the matching relationship between the control parameter configuration and the effect value in the execution record.

6. The IoT edge control system applied to sewage treatment according to claim 1 is characterized in that: Initialize the device by following the steps below: Start the edge computing device and connect it to the cloud through the smart gateway relay to obtain cloud services; The edge controller reads the initial sequence of sensor measurements, uploads them to the cloud as initial data, and continues to measure data; The cloud uses the priority recommendation module to query the historical measurement value sequence that meets the matching threshold from the historical time series database based on the initial measurement value sequence using the optimal matching algorithm; Setting a cloud Kalman filter in the cloud and inputting a historical measurement value sequence into the cloud Kalman filter; Initialize the cloud Kalman filter in the cloud and record the initialized filter parameters; The cloud sends the filter parameters to the edge computing device, and the edge controller sets the local Kalman filter and controls the operation of the local Kalman filter according to the filter parameters; The local Kalman filter is continuously run, the corresponding filter parameters are output and uploaded to the historical time series database for storage.

7. The IoT edge control system applied to sewage treatment according to claim 1 is characterized in that: The priority recommendation module uses a moving window average method to reversely search or screens the time series through the mean / median to set the optimal matching algorithm.

8. The IoT edge control system applied to sewage treatment according to claim 1 is characterized in that: The intelligent gateway also performs hidden settings for the wireless local area network and authenticates the edge computing device; the edge device is connected to the hidden wireless local area network through an encrypted channel.

Images