CN-121980338-A - Multistage AO sewage treatment real-time supervision and fault diagnosis system based on thing networking

CN121980338ACN 121980338 ACN121980338 ACN 121980338ACN-121980338-A

Abstract

The invention discloses a multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the Internet of things, which relates to the field of the Internet of things and comprises the steps of collecting multi-dimensional data of an anaerobic tank, an anoxic tank, an aerobic tank and a secondary sedimentation tank in real time through a multi-stage AO process monitoring module, encrypting the multi-dimensional data through an Internet of things data transmission module and uploading the encrypted multi-dimensional data to a cloud; the edge calculation module performs noise filtering and data complement to generate a preliminary regulation instruction, the fault diagnosis module combines a deep convolutional neural network and a gradient lifting tree algorithm to realize accurate fault classification and tracing, parameters such as aeration, stirring, reflux ratio and the like are automatically adjusted according to diagnosis results, and an alarm is pushed through a multi-stage early warning mechanism, and the cloud adopts distributed storage to support multidimensional retrieval and optimizes model parameters based on historical data. The method has the advantages that accurate fault diagnosis and collaborative regulation are achieved through edge calculation and intelligent algorithm, a closed loop is formed by combining cloud management and remote operation and maintenance, process stability and fault processing efficiency are improved, and operation and maintenance cost is reduced.

Inventors

WANG SHUNHE
LI KAI
HU SIQI
WU ZUSHENG

Assignees

宇星环保工程有限公司

Dates

Publication Date: 20260505
Application Date: 20260106

Claims (9)

1. Multistage AO sewage treatment real-time supervision and fault diagnosis system based on thing networking, its characterized in that includes: The multi-stage AO process monitoring module is arranged in an anaerobic tank, an anoxic tank, an aerobic tank and a secondary sedimentation tank of the multi-stage AO sewage treatment process and comprises a multi-dimensional sensing assembly and a self-adaptive sampling unit; the data transmission module of the Internet of things adopts an edge node-gateway-cloud three-level transmission architecture, gathers and primarily filters the data of each sensor nearby, encrypts the transmission data by adopting an encryption algorithm, and converts a data private protocol into an MQTT standard protocol; the edge calculation data processing module is used for removing random noise in the sensor data, carrying out intelligent complementation on the missing data based on a material balance principle of a multi-stage AO process, and generating a preliminary regulation and control instruction; The fault diagnosis module is used for extracting characteristics of the preprocessing data and the historical operation data through a deep convolutional neural network algorithm, precisely classifying fault types by combining a gradient lifting tree algorithm, constructing a correlation model of faults and reasons, and generating a diagnosis result; The regulation and control execution module is connected with execution equipment of the multi-stage AO process and is used for carrying out aeration regulation and control, stirring regulation and control, reflux ratio regulation and mud discharge regulation and control based on the preliminary regulation and control instruction and the diagnosis result; The early warning pushing module is used for setting multi-level early warning based on the diagnosis result of the fault diagnosis module, wherein each level corresponds to different early warning threshold values and processing priorities, and pushing early warning information in a plurality of pushing modes; The cloud data management module is used for safely storing the total data by adopting a distributed storage architecture, carrying out multidimensional retrieval based on time, process and fault types, constructing a historical data mining model, and optimizing parameters and regulation strategy thresholds of the fault diagnosis model by analyzing historical operation data and fault processing data; And the remote operation and maintenance module is used for checking the operation state, the fault diagnosis result and the early warning information of each module in real time through the Web end and the APP end, remotely issuing operation and maintenance instructions and generating an equipment maintenance plan based on equipment data.
2. The internet of things-based multi-stage AO sewage treatment real-time monitoring and fault diagnosis system according to claim 1, wherein the multi-stage AO process monitoring module specifically comprises: The multi-dimensional sensing assembly unit comprises a dissolved oxygen sensor, a COD sensor, an ammonia nitrogen sensor, a total nitrogen sensor, a sludge concentration sensor, a pH sensor and a tank body liquid level sensor, wherein each sensor is provided with a PTFE composite antifouling blocking membrane; The self-adaptive sampling unit integrates a flow sensor and an electric lifting sampling mechanism, a fluctuation threshold is set based on real-time flow data, the sampling frequency is automatically lifted when the flow fluctuation exceeds +/-10% or the water quality parameter fluctuation exceeds +/-5%, and the electric lifting sampling rod mechanism is driven by a stepping motor to position the sampling depth.
3. The multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the internet of things according to claim 1, wherein the data transmission module of the internet of things specifically comprises: the edge node communication unit adopts a LoRa and NB-IoT dual-mode integrated design to collect nearby data of each sensor and perform preliminary filtering, and automatically switches NB-IoT signals when the LoRa signals are interrupted; the gateway processing unit is used for carrying out hardware encryption on the data uploaded by the edge node, dynamically generating a secret key through the unique equipment identifier and converting the sensor private protocol into an MQTT standard protocol in real time; and the cloud communication unit adopts 4G/5G and optical fiber dual-mode redundancy transmission, performs integrity check on transmission data based on a CRC-32 algorithm, and triggers a retransmission mechanism if the check fails.
4. The multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the internet of things according to claim 1, wherein the edge calculation data processing module specifically comprises: The data preprocessing unit is used for removing random noise in sensor data by adopting a sliding window filtering algorithm, dynamically adjusting the window length according to data fluctuation, correcting systematic errors by comparing standard reference values of calibration curve parameters of the sensors of all types in real time, and completing data based on parameters of adjacent processing units; The edge decision operation unit is used for constructing an edge decision model based on a lightweight neural network algorithm and removing a redundant network layer aiming at an edge end computing power limited scene optimization model structure; And the regulation and control instruction generation unit is internally provided with a fault symptom identification rule base, inputs the preprocessed real-time data into an edge decision model, acquires simple fault symptoms, and generates a preliminary regulation and control instruction by matching the fault symptom identification rule base.
5. The multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the internet of things according to claim 1, wherein the fault diagnosis module specifically comprises: The feature extraction unit is used for extracting features of the preprocessed data transmitted by the edge calculation data processing module, wherein the features comprise a water quality parameter fluctuation coefficient, an equipment operation parameter deviation value and a process unit response delay, interference features are removed through signal smoothing processing, and key features which are strongly related to faults are reserved based on a mutual information entropy principle; the fault classification and positioning unit is used for realizing fault classification by adopting a deep convolutional neural network and gradient lifting tree fusion model, constructing a fault-cause correlation network through a knowledge graph, positioning a fault source and generating a diagnosis result containing fault types, influence ranges and processing steps; and the diagnosis model iteration and storage unit is used for caching the recent fault data and diagnosis results for 1 month, receiving model parameters optimized by the cloud based on historical data in real time, and updating the model increment.
6. The multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the internet of things according to claim 1, wherein the regulation and control execution module specifically comprises: an aeration regulation and control unit, which adopts an industrial vector frequency converter and a high-efficiency aeration head array, receives regulation and control instructions through 4-20mA analog signals, deploys independent aeration branches in each aerobic tank partition, and controls aeration quantity of each area through branch electromagnetic valves; The stirring regulation and control unit selects a submersible low-speed plug flow stirrer, and the stirring paddle adopts a propeller type structure and controls the plug flow intensity by regulating the stirring rotating speed; The reflux ratio regulating and controlling unit integrates an electromagnetic flow sensor and a variable-frequency reflux pump, the flow sensor collects reflux flow of nitrified liquid and sludge in real time, and the reflux pump regulates the rotation speed of the pump to realize dynamic regulation and control of the reflux ratio; And the sludge discharge regulation and control unit triggers sludge discharge action through real-time data of the sludge concentration sensor, increases the opening degree when the sludge concentration exceeds a set threshold value, and decreases the opening degree when the sludge concentration is lower than the threshold value.
7. The multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the internet of things according to claim 1, wherein the early warning pushing module specifically comprises: the multi-stage early warning judging unit is used for judging the primary, secondary and tertiary early warning grades by combining the operation threshold standard of each unit of the multi-stage AO process based on the diagnosis result of the fault diagnosis module and the parameter exceeding degree of the edge calculation data processing module; the multi-channel pushing execution unit constructs a multi-channel redundancy pushing framework integrating GSM short messages, 4G communication, audible and visual alarm and industrial Ethernet, and self-defines pushing content and pushing frequency based on the received object identity group; and the early warning ledger and feedback unit is used for recording early warning generation time, grade, triggering reason, pushing object, receiving state and processing result in real time to form a traceable early warning ledger, and the operation and maintenance personnel feeds back early warning processing progress through the APP.
8. The multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the internet of things according to claim 1, wherein the cloud data management module specifically comprises: The distributed data storage unit adopts a mixed architecture of distributed file storage, a relational database and a time sequence database, wherein the distributed file storage adopts SSD and mechanical hard disk layered design, the relational database stores equipment information and operation and maintenance ledger structured data, and the time sequence database stores water quality and process parameter time sequence data; The data processing and searching unit optimizes the data quality by eliminating abnormal values, standardizing the data format and repeating the data deduplication operation, and the data searching supports multidimensional combination searching based on time, process units, data types and fault types; and the model optimizing unit is used for optimizing the parameter threshold value of the fault diagnosis module and the regulation strategy of the regulation execution module by analyzing the historical operation data and the fault processing record, generating an optimized parameter packet and pushing the optimized parameter packet to the edge.
9. The multi-stage AO sewage treatment real-time monitoring and fault diagnosis system based on the internet of things according to claim 1, wherein the remote operation and maintenance module specifically comprises: the instruction interaction unit is internally provided with a permission grading management module, three-level permissions are divided according to an administrator, an operation and maintenance person and an operator, only the administrator can issue an operation and maintenance instruction, the execution state is fed back in real time after the instruction is issued, and a retry mechanism and an abnormal alarm are triggered when the execution fails; the equipment maintenance management unit pre-stores model parameters, maintenance period and spare part information of the aeration fan and the reflux pump equipment, and generates a periodic maintenance plan based on equipment operation time length, failure frequency and load rate data; And the remote upgrading and operation and maintenance ensuring unit adopts a differential upgrading technology to remotely upgrade the firmware of the core module, only transmits the difference data of the upgrading packet, and is internally provided with a local operation and maintenance interface to carry out field debugging and emergency fault processing.

Description

Multistage AO sewage treatment real-time supervision and fault diagnosis system based on thing networking Technical Field The invention relates to the field of Internet of things, in particular to a multistage AO sewage treatment real-time monitoring and fault diagnosis system based on the Internet of things. Background Along with the rapid development of industrialization and city, sewage treatment becomes a key link of environmental protection and water resource recycling. The multistage AO (anaerobic-aerobic) process is widely applied to sewage treatment plants due to the high-efficiency denitrification and dephosphorization capability, but the operation process involves complex biochemical reaction and multiple control parameters, is easily influenced by factors such as water inlet load, dissolved oxygen, sludge concentration and the like, and causes fluctuation of treatment efficiency and even system failure. The current sewage treatment monitoring and regulating system on the market is difficult to match with the accurate and intelligent operation demands. Most systems have single monitoring dimension, lack of self-adaptive sampling mechanism, weak impact resistance when facing water quality and water quantity fluctuation, and easy occurrence of monitoring data deviation. The data transmission adopts a single architecture, is easy to be interfered to cause packet loss and delay, has insufficient encryption and protocol conversion capability, and has poor data security and compatibility. In the aspects of data processing and fault diagnosis, the method lacks of real-time preprocessing capability of edge calculation, relies on traditional algorithms or manual judgment, is inaccurate in fault feature extraction, is fuzzy in classification and positioning, and is difficult to pre-judge fault symptoms in advance. Most of regulation and control execution is rough unified control, and the capacity of regional cooperative regulation and control is lacked, and response is lagged. Meanwhile, operation and maintenance management relies on-site manual inspection, remote control and intelligent maintenance capability are lost, multi-platform data are independently split, overall process overall analysis and model iterative optimization cannot be achieved, overall operation efficiency is low, operation and maintenance cost is high, and environment-friendly standard risk is high. Disclosure of Invention In order to perfect the existing system, the system for monitoring and diagnosing the multi-stage AO sewage treatment in real time based on the Internet of things is provided, the system realizes the precise monitoring and safe data transmission of the whole multi-stage AO sewage treatment process by means of the Internet of things, achieves precise diagnosis and cooperative regulation and control of faults through edge calculation and intelligent algorithms, combines cloud management and remote operation and maintenance to form a closed loop, greatly improves the process stability and fault treatment efficiency, and reduces operation and maintenance cost. In order to achieve the above purpose, the invention adopts the following technical scheme: The multi-stage AO process monitoring module is arranged in an anaerobic tank, an anoxic tank, an aerobic tank and a secondary sedimentation tank of the multi-stage AO sewage treatment process and comprises a multi-dimensional sensing assembly and a self-adaptive sampling unit; the data transmission module of the Internet of things adopts an edge node-gateway-cloud three-level transmission architecture, gathers and primarily filters the data of each sensor nearby, encrypts the transmission data by adopting an encryption algorithm, and converts a data private protocol into an MQTT standard protocol; the edge calculation data processing module is used for removing random noise in the sensor data, carrying out intelligent complementation on the missing data based on a material balance principle of a multi-stage AO process, and generating a preliminary regulation and control instruction; The fault diagnosis module is used for extracting characteristics of the preprocessing data and the historical operation data through a deep convolutional neural network algorithm, precisely classifying fault types by combining a gradient lifting tree algorithm, constructing a correlation model of faults and reasons, and generating a diagnosis result; The regulation and control execution module is connected with execution equipment of the multi-stage AO process and is used for carrying out aeration regulation and control, stirring regulation and control, reflux ratio regulation and mud discharge regulation and control based on the preliminary regulation and control instruction and the diagnosis result; The early warning pushing module is used for setting multi-level early warning based on the diagnosis result of the fault diagnosis module, wherein each level corresponds to different early warnin