CN-121974480-A - Aeration control, system, equipment and storage medium for sewage treatment process

CN121974480ACN 121974480 ACN121974480 ACN 121974480ACN-121974480-A

Abstract

The invention provides an aeration control method of a sewage treatment process, which comprises the steps of obtaining real-time process parameters of the sewage treatment process, inputting the real-time process parameters into a pre-trained microbial oxygen consumption prediction model to obtain a future oxygen consumption predicted value, inputting the real-time process parameters and the future oxygen consumption predicted value into a pre-trained decision control model, generating an aeration control instruction based on a multi-objective optimization function by the decision control model, and sending the aeration control instruction to control aeration equipment. The method comprises the steps of inputting acquired real-time process parameters into a microbial oxygen consumption prediction module to obtain a future oxygen consumption prediction value, performing prospective prediction on the future oxygen consumption, generating an aeration control instruction based on an optimization target through a decision control model to realize intelligent decision, and controlling aeration equipment to complete a control closed loop through a sending instruction. The invention effectively solves the technical problems of disjoint control logic and the real aerobic demand of microorganisms and insufficient control precision in the traditional aeration control method.

Inventors

Pi Zhiping
CHEN SHIJIE
HUANG CHI
SHEN XU
LUO ZILIANG

Assignees

昕彤赋能(长沙)人工智能行业应用系统有限公司

Dates

Publication Date: 20260505
Application Date: 20260408

Claims (10)

1. An aeration control method for a sewage treatment process, comprising the steps of: acquiring real-time technological parameters of a sewage treatment process; Inputting the real-time process parameters into a pre-trained microbial oxygen consumption prediction model to obtain a future oxygen consumption prediction value; inputting the real-time process parameters and the future oxygen consumption predicted value into a pre-trained decision control model to obtain an aeration control instruction, wherein the decision control model is obtained by training by taking a multi-objective optimization function as a training target; And sending the aeration control instruction to control the aeration equipment.
2. An aeration control method according to claim 1, wherein the real-time process parameters include influent water quality data, biochemical tank state data, aeration equipment operation state data, and time-of-use electricity price data.
3. The aeration control method according to claim 2, wherein the microbial oxygen consumption prediction model is a deep learning model based on a causal discovery algorithm; Inputting the real-time process parameters into a pre-trained microbial oxygen consumption prediction model to obtain a future oxygen consumption predicted value, wherein the method comprises the following steps of: Identifying key process characteristics which are causally related to the oxygen consumption rate of microorganisms in the real-time process parameters from a pre-generated causal map through a causal discovery algorithm in the deep learning model; based on the key process characteristics, estimating and obtaining a real-time oxygen consumption rate value; and obtaining the future oxygen consumption predicted value based on the real-time oxygen consumption rate value.
4. An aeration control method according to claim 3, wherein the decision control model is a strategy network based on a deep reinforcement learning algorithm; the multi-objective optimization function is a punishment function and is formed by weighted summation of an energy consumption punishment item, an oxygen demand deviation punishment item and a stability punishment item, wherein the energy consumption punishment item is obtained by calculation based on time-sharing electricity prices in the real-time process parameters; inputting the real-time technological parameter and the future oxygen consumption predicted value into a pre-trained decision control model to obtain an aeration control instruction, wherein the method comprises the following steps of: Inputting the real-time process parameters and the future oxygen consumption predicted value into the strategy network, wherein the strategy network outputs aeration control instructions which minimize the punishment function value.
5. An aeration control method according to claim 4, wherein said aeration control command includes a blower group control command, a valve opening control command, and a manifold pressure control command.
6. An aeration control method according to claim 3, wherein said inputting said real-time process parameters into a pre-trained microbial oxygen consumption prediction model to obtain future oxygen consumption predictions, further comprises: monitoring the sensor state corresponding to the key process characteristics in real time; when the sensor is detected to be faulty, determining that an alternative process feature which is causally related to the key process feature corresponding to the faulty sensor exists based on the causal map; Based on the alternative process characteristics, estimating to obtain a real-time oxygen consumption rate value; and obtaining the future oxygen consumption predicted value based on the real-time oxygen consumption rate value.
7. An aeration control method according to claim 1, characterized by further comprising, after the aeration control instruction is sent, after the aeration equipment is controlled: acquiring real-time technological parameters after executing the aeration control instruction to form empirical data; And carrying out iterative training on the microbial oxygen consumption prediction model and/or the decision control model based on the empirical data so as to realize continuous optimization of a control strategy.
8. An aeration control system for a wastewater treatment process, comprising: The acquisition module is used for acquiring real-time process parameters of the sewage treatment process; the prediction module is used for inputting the real-time process parameters into a pre-trained microbial oxygen consumption prediction model to obtain a future oxygen consumption predicted value; The instruction generation module is used for inputting the real-time process parameters and the future oxygen consumption predicted value into a pre-trained decision control model to obtain an aeration control instruction, wherein the decision control model is obtained by training by taking a multi-objective optimization function as a training target; And the control module is used for sending the aeration control instruction and controlling the aeration equipment.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the aeration control method according to any one of claims 1-7 when the computer program is executed.
10. A computer-readable storage medium storing a computer program which, when executed by a processor, implements the steps of the aeration control method according to any one of claims 1 to 7.

Description

Aeration control, system, equipment and storage medium for sewage treatment process Technical Field The invention relates to the technical field of sewage treatment, in particular to aeration control, a system, equipment and a storage medium of a sewage treatment process. Background In the sewage treatment process, an aeration system is a key link of aerobic biochemical treatment, and the main function of the aeration system is to provide sufficient dissolved oxygen for activated sludge microorganisms so as to meet the aerobic requirements of organic matter degradation and nitration reaction. However, the aeration system is also the most important energy consumption unit of the sewage treatment plant, and the electricity consumption of the aeration system is usually more than 50% of the total electricity consumption of the whole plant. Currently, the main stream aeration control method mainly adopts PID control based on fixed dissolved oxygen (Dissolved Oxygen, DO) set values. However, this control approach suffers from the technical disadvantage that, first, the control logic is essentially decoupled from the process. PID controllers are dedicated to maintaining the dissolved oxygen concentration at a fixed value set manually, but the true oxygen demand of microorganisms fluctuates drastically with factors such as feed water contaminant load, water temperature, sludge activity, etc. The DO set value is fixed, so that the excessive aeration causes electric energy waste at low load, and the insufficient aeration causes incomplete nitrification and out-of-water ammonia nitrogen exceeding standard at high load. Second, there is a severe hysteresis in the control. DO itself has control challenges of hysteresis, complexity, and nonlinearity. PID control adjusts aeration according to the current DO value, but the current DO value reflects the adjustment result before 15 minutes, and the hysteresis can lead to continuous oscillation of the DO value, so that the control accuracy is generally lower. Third, it is difficult to achieve cooperative control of multiple devices. In traditional control, variable frequency regulation of a blower and opening control of each valve on an air pipe network are often cut, so that the blower is often operated at an inefficient working point, or the air quantity distribution of different aeration areas is unreasonable, and the overall aeration efficiency is low. Fourth, relies on expensive hardware and lacks fault tolerance capability. Professional instruments for online measurement of the oxygen consumption rate (Oxygen Uptake Rate, OUR) of activated sludge are extremely expensive, resulting in failure of most sewage treatment plants. Meanwhile, when the sensor fails, the traditional system usually gives an alarm and stops the machine directly, and an intelligent fault-tolerant mechanism is lacked. Therefore, it is needed to provide an aeration control method for realizing the cooperative control of multiple devices based on the real aerobic demands of the microorganisms with accurate perception and simultaneously having intelligent fault tolerance. Disclosure of Invention The method and the system realize multidimensional state sensing by acquiring real-time process parameters, lay a data foundation for accurate control, obtain future oxygen consumption predicted values through a microbial oxygen consumption prediction model, conduct future oxygen consumption prediction, generate aeration control instructions based on optimization targets through a decision control model to realize intelligent decision, control aeration equipment to complete control closed loop through sending the instructions, and guarantee accurate execution of strategies. The invention effectively solves the technical problems of disjoint control logic and real aerobic demand of microorganisms and lack of foresight in the traditional aeration control method. The invention aims to provide an aeration control method of a sewage treatment process; The technical scheme provided by the invention is as follows: an aeration control method of a sewage treatment process, comprising: acquiring real-time technological parameters of a sewage treatment process; Inputting the real-time process parameters into a pre-trained microbial oxygen consumption prediction model to obtain a future oxygen consumption prediction value; Inputting real-time technological parameters and future oxygen consumption predicted values into a pre-trained decision control model to obtain an aeration control instruction, wherein the decision control model is obtained by taking a multi-objective optimization function as a training target; And sending the aeration control instruction to control the aeration equipment. Preferably, the real-time process parameters comprise water quality data of the inlet water, biochemical pool state data, operation state data of the aeration equipment and time-of-use electricity price data. Preferably, the microbial o