Coal chemical heavy oil sewage treatment system based on Internet of Things

Abstract

The invention discloses a coal chemical industry heavy oil sewage treatment system based on the Internet of Things, and relates to the technical field of coal chemical industry sewage treatment.The coal chemical industry heavy oil sewage treatment system achieves real-time monitoring of sewage and membrane surface pollutant attachment conditions through an Internet of Things sensor, processes data through a machine learning algorithm, evaluates the current membrane pollution degree, and improves the membrane pollution degree. According to the invention, pollution is accurately identified, a response is made immediately when pollution occurs, the cleaning period and the cleaning strength are intelligently adjusted by dynamically analyzing the pollution degree and the sewage treatment load, the starting time of cleaning is determined according to the actual pollution condition, excessive cleaning is avoided, the cleaning period can be accelerated when needed, and the resource utilization efficiency is improved. The membrane pollution trend is predicted through the long short-term memory network LSTM, the potential pollution problem is recognized in advance, the system is helped to take prevention measures before pollution deterioration, and the problem of system shutdown or treatment capacity reduction caused by membrane blockage is effectively avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of coal chemical wastewater treatment, and in particular to a coal chemical heavy oil wastewater treatment system based on the Internet of Things. Background Art

[0002] As an important part of the energy industry, coal chemical industry produces wastewater with high concentration and heavy pollution, which has become a difficulty in environmental protection. In order to solve this problem, membrane technology for deep treatment of coal chemical wastewater came into being and played an important role in the field of wastewater treatment. This technology uses membrane separation process to deeply treat pollutants in coal chemical wastewater to improve water quality and achieve standard discharge, thereby effectively reducing environmental pollution.

[0003] Coal chemical wastewater contains high concentrations of oil, organic matter and suspended particles. These substances are easy to adhere to the membrane surface, which can easily cause membrane fouling, significantly reduce the membrane flux and separation efficiency and increase operating costs. Membrane fouling occurs frequently, and existing treatment methods mainly rely on regular chemical or physical cleaning, which makes it difficult to achieve timely warning and control. The lack of flexibility often leads to excessive or insufficient cleaning, increases membrane loss and affects system stability.

[0004] In response to the problem of fixed periodic cleaning, some traditional treatment systems use manual adjustment of the cleaning cycle and intensity, and manually determine when to clean based on changes in membrane flux and effluent water quality. However, this is prone to misjudgment or delayed response, and it is difficult to track the membrane contamination in real time. Moreover, when membrane contamination rapidly increases or water quality fluctuates greatly, manual adjustments are difficult to respond in a timely manner, and automated management cannot be achieved. Some treatment systems use automatic cleaning methods, setting regular chemical cleaning and backwashing procedures to restore filtration capacity by removing oil, organic matter and particulate matter from the membrane surface. However, most cleaning procedures are also regular and cannot be arranged according to the degree of contamination, which will also lead to excessive or insufficient cleaning. Excessive cleaning increases operating costs, while insufficient cleaning will aggravate membrane contamination, reduce treatment efficiency and shorten membrane life. Therefore, there is an urgent need for a coal chemical heavy oil wastewater treatment system based on the Internet of Things to solve such problems. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a coal chemical heavy oil wastewater treatment system based on the Internet of Things to solve the problems that traditional wastewater treatment systems cannot monitor the contamination of membranes in real time, it is difficult to flexibly adjust the cleaning cycle and operating conditions according to the actual degree of contamination, the response lag of the cleaning and maintenance processes is strong, and the fixed cycle cleaning method is difficult to accurately control the cleaning timing, resulting in excessive use of water resources and cleaning agents, and frequent cleaning may accelerate the loss of membranes and increase the cost of equipment maintenance and replacement.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] On the one hand, the present invention provides a coal chemical heavy oil wastewater treatment system based on the Internet of Things, which comprises:

[0009] The central control system is responsible for receiving data from IoT sensors and membrane contamination assessment results from the IoT platform, issuing operation instructions based on the membrane contamination assessment results, and adjusting the operating status of the membrane separation system, cleaning management unit, and pressure and flow control unit;

[0010] The central control system includes:

[0011] The cleaning management unit is responsible for adjusting the amount of cleaning agent, cleaning frequency and flushing water flow rate;

[0012] The pressure and flow control unit is responsible for adjusting the inlet and outlet water pressure and flow of the membrane separation system;

[0013] The membrane separation system is responsible for the sewage filtration process. It uses membrane technology to remove suspended solids, oils and macromolecular organic matter in heavy oil sewage. The sewage is driven through the membrane by the inlet and outlet water pressure and flow rate for separation and purification.

[0014] The IoT platform is responsible for receiving real-time data from IoT sensors, evaluating the degree of membrane contamination based on the real-time data, and transmitting the results of the membrane contamination evaluation to the central control system.

[0015] IoT sensors are responsible for collecting real-time data in the sewage treatment system, including the membrane inlet and outlet water pressure difference, flux, flow rate and pollutant concentration, and sending the real-time data to the IoT platform.

[0016] On the other hand, the present invention provides a method for treating coal chemical heavy oil wastewater based on the Internet of Things, comprising:

[0017] Step S1, data collection and monitoring,

[0018] IoT sensors monitor membrane-related real-time data in the sewage treatment system in real time and upload the real-time data to the central control system and IoT platform;

[0019] Step S2, membrane fouling trend identification,

[0020] Use big data analysis and machine learning algorithms on the IoT platform to evaluate membrane contamination and predict membrane contamination trends. Membrane contamination includes wastewater composition and pollutant attachment. Based on wastewater composition and pollutant attachment, the current membrane contamination level is evaluated.

[0021] Step S3, optimization of cleaning cycle and intensity scheduling,

[0022] Based on the current membrane contamination degree generated in step S2, the cleaning cycle and intensity of the membrane separation system are adjusted, the specific time point for starting the cleaning operation is determined, and the amount of cleaning agent used and the flow rate of flushing water are adjusted according to the sewage treatment load and the membrane contamination degree;

[0023] Step S4, dynamic pressure and flow control,

[0024] According to the real-time data in step S1 and the cleaning plan in step S3, the central control system dynamically adjusts the inlet and outlet water pressures and flow rates of the membrane separation system, and automatically adjusts the water pump pressure when the sensor detects a decrease in flux or an increase in pressure difference.

[0025] Furthermore, in step S1, the real-time data includes the inlet and outlet water pressure difference, flux, flow rate and pollutant concentration.

[0026] Furthermore, in step S2, the current membrane fouling degree is evaluated:

[0027] Obtain pollutant concentration data from IoT sensors, including:

[0028] Sewage composition: suspended matter concentration C SS , Chemical oxygen demand C COD , Oil pollutant concentration C oil , Salt concentration C salt ;

[0029] Pollutant adhesion: The amount of organic matter attached to the membrane surface A organic 、Inorganic matter A inorganic 、Amount of oily substances attached oil ;

[0030] C SS , reflects the content of particulate matter in sewage, C COD , reflects the concentration of organic matter in sewage, C oil , chemical fouling affecting the membrane, C salt , easy to cause inorganic matter to adhere, A organic , indicating the degree of contamination of the membrane surface by organic pollutants, A inorganic , reflecting the situation of inorganic pollutants blocking the membrane pores, A oil , represents the degree of contamination of the membrane surface by oily substances;

[0031] Extract relevant features of real-time data and calculate the change rate and acceleration of each feature:

[0032] The rate of change, or first-order derivative, is expressed as They respectively represent the changing rates of the data corresponding to the sewage composition and the pollutant attachment conditions;

[0033] The changing acceleration, or second-order derivative, is expressed as They respectively represent the acceleration trends of suspended matter concentration, chemical oxygen demand, the amount of organic matter attached to the membrane surface, and the amount of oil attached.

[0034] Furthermore, in step S2, the method for evaluating the current membrane contamination degree also includes:

[0035] Calculate the attachment rate of pollutants: R attachment = k·(C in -C out )·Q, where R attachment is the attachment rate of pollutants, which indicates the accumulation rate of pollutants on the membrane surface per unit time, C in is the pollutant concentration in the influent, C out is the pollutant concentration in the effluent, reflecting the removal effect of the sewage after passing through the membrane, Q is the sewage flow rate, indicating the amount of sewage flowing through the membrane per unit time, and k is the attachment rate coefficient, which is adjusted according to the membrane characteristics and the type of pollutants;

[0036] Based on the attachment rate, the accumulation of membrane pollutants and the degree of membrane surface pollution are calculated:

[0037] Among them, A total (t) is the total amount of pollutants attached to the membrane surface at time t, indicating the accumulation of pollutants on the membrane surface, R attachment (t) is the pollutant attachment rate, t is time, and the integral represents the cumulative effect from the initial moment to the current moment. The total amount of pollutants on the membrane surface can be calculated through this formula.

[0038] Furthermore, in step S2, the membrane fouling trend is predicted as follows:

[0039] The long short-term memory network LSTM is used to capture the long-term and short-term dependencies in the time series and identify the impact of sewage composition and pollutant attachment on membrane fouling. For each moment t, the LSTM model consists of:

[0040] Forget Gate t :f t =σ(W f ·[h t-1 ,x t ]+b f ),

[0041] Input Gate i t :i t =σ(W i ·[h t-1 ,x t ]+b i ),

[0042] Candidate memory cells

[0043] Update memory unit C t :

[0044] Output gate o t :o t =σ(W o ·[h t-1 ,x t ]+b o ), h t =o t tanh(C t );

[0045] Among them, x t Represent input characteristics, including sewage composition and pollutant attachment data characteristics, as well as change rate, change acceleration, attachment rate and accumulation per unit time on the membrane surface;

[0046] h t-1 Indicates the hidden state at the previous moment; W f ,W i ,W C ,W o represents the weight matrix, b f ,b i ,b C ,b o is the bias, σ(·) is the Sigmoid activation function, and tanh(·) is the hyperbolic tangent function.

[0047] Furthermore, in step S2, the method for predicting the membrane fouling trend also includes:

[0048] Through the recursive calculation of the LSTM network, the model finally outputs the predicted value of membrane pollution and predicts the future membrane pollution trend:

[0049] in, is the predicted value of membrane fouling at the future time t+T, indicating the predicted degree of fouling, W o represents the weight matrix of the output layer, h t represents the hidden layer state at the current moment, reflecting the combined influence of pollutant concentration and adhesion, σ(·) is the Sigmoid activation function, b o Represents the bias of the output layer. Through the LSTM model, the membrane pollution trend prediction at the future time t+T is generated to provide a basis for cleaning scheduling and pollution warning.

[0050] Furthermore, in step S3, the wastewater treatment load refers to the amount of wastewater and the concentration of pollutants treated by the membrane in the membrane separation system.

[0051] Furthermore, the cleaning cycle and intensity scheduling method in step S3 is:

[0052] According to the membrane contamination degree calculated in step S2 and the current sewage treatment load, it is determined whether the cleaning operation needs to be started through a preset threshold value;

[0053] According to the degree of contamination, the membrane contamination is divided into mild, moderate and severe, and the corresponding cleaning cycle is set;

[0054] When the membrane fouling level reaches a preset threshold, cleaning is initiated.

[0055] Furthermore, in step S4, the dynamic pressure and flow rate control method is:

[0056] The sensor monitors the water inlet pressure P in real time in (t), water outlet pressure P out (t), membrane pressure difference ΔP(t) = P in (t)-P out (t), flux J(t) and flow rate Q(t);

[0057] Calculate the pressure difference change rate and flux change rate in the membrane system:

[0058] in, represents the rate of change of pressure difference with time t, C in and C out They represent the pollutant concentrations of the inlet and outlet water, respectively, Q(t) represents the sewage flow rate, which affects the change of the pressure difference, and k1 represents the empirical coefficient, that is, the contribution of the pollutant concentration difference to the pressure difference;

[0059] in, is the rate of change of flux J(t), indicating the change of filtration efficiency, P in (t) is the inlet pressure, which pushes the fluid through the membrane. ΔP(t) is the pressure difference. An increase in the pressure difference reduces the flux. A fouling (t) is the amount of membrane pollutant attachment. As the pollutants attach, the flux gradually decreases. k2 and k3 are empirical coefficients that reflect the effects of inlet pressure and pollutants on flux changes.

[0060] When the pressure difference ΔP(t) increases or the flux J(t) decreases, the pressure and flow of the water pump are automatically adjusted. The water pump pressure regulation formula is:

[0061] Among them, P pump (t) is the pressure of the pump at time t, P pump (t-1) is the pump pressure at the previous moment t-1, is the rate of change of pressure difference. When the pressure difference increases, the pump pressure needs to be increased. is the rate of change of flux. When the flux decreases, the pump pressure also needs to be increased. α and β are adjustment coefficients, which are used to balance the effects of pressure difference and flux change on the pump pressure adjustment.

[0062] Water flow regulation formula: Where, Q(t) is the current water flow, Q set is the set ideal flow rate, ΔP(t) is the real-time pressure difference, ΔP max is the maximum allowable pressure difference. When the pressure difference ΔP(t) is close to ΔP max When the pressure difference increases, the flow rate decreases gradually, and γ is the adjustment coefficient, which controls the degree of flow reduction. When the pressure difference increases, the flow rate is automatically reduced to reduce the load on the membrane, thereby extending the service life of the membrane.

[0063] The beneficial effects of the present invention are:

[0064] The present invention realizes real-time monitoring of sewage composition and the attachment of pollutants on the membrane surface through Internet of Things sensors, and uses machine learning algorithms to process data, evaluate the current degree of membrane contamination, accurately identify contamination, and respond immediately when contamination occurs, avoiding membrane blockage and efficiency reduction problems caused by delays in traditional methods.

[0065] The present invention dynamically analyzes the pollution degree and sewage treatment load, intelligently adjusts the cleaning cycle and cleaning intensity, determines the start time of cleaning according to the actual pollution situation, avoids excessive cleaning, and speeds up the cleaning cycle when necessary, thereby improving resource utilization efficiency, reducing the waste of cleaning agents and water resources, and extending the service life of the membrane.

[0066] The present invention predicts the membrane pollution trend through the long short-term memory network LSTM, identifies potential pollution problems in advance, and helps the system take preventive measures before the pollution worsens. The active prediction capability enhances the adaptability of the system and can effectively avoid the system shutdown or reduced processing capacity caused by membrane blockage.

[0067] The present invention monitors the pressure difference and flux changes of the membrane system in real time and dynamically adjusts the water pump pressure and flow rate, thereby effectively alleviating the negative impact of membrane pollution on system performance and reducing the risk of membrane damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0069] Figure 1This is a schematic diagram of the process of treating heavy oil wastewater from coal chemical industry of the present invention;

[0070] Figure 2 It is a structural schematic diagram of the coal chemical heavy oil wastewater treatment system of the present invention. DETAILED DESCRIPTION

[0071] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0072] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0073] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0074] Example 1, reference Figure 2 , this embodiment provides a coal chemical heavy oil wastewater treatment system based on the Internet of Things, including:

[0075] The central control system is responsible for receiving data from IoT sensors and membrane contamination assessment results from the IoT platform, issuing operation instructions based on the membrane contamination assessment results, and adjusting the operating status of the membrane separation system, cleaning management unit, and pressure and flow control unit;

[0076] The central control system includes:

[0077] The cleaning management unit is responsible for adjusting the amount of cleaning agent, cleaning frequency and flushing water flow rate;

[0078] The pressure and flow control unit is responsible for adjusting the inlet and outlet water pressure and flow of the membrane separation system;

[0079] The membrane separation system is responsible for the sewage filtration process. It uses membrane technology to remove suspended solids, oils and macromolecular organic matter in heavy oil sewage. The sewage is driven through the membrane by the inlet and outlet water pressure and flow rate for separation and purification.

[0080] The IoT platform is responsible for receiving real-time data from IoT sensors, evaluating the degree of membrane contamination based on the real-time data, and transmitting the results of the membrane contamination evaluation to the central control system.

[0081] IoT sensors are responsible for collecting real-time data in the sewage treatment system, including the membrane inlet and outlet water pressure difference, flux, flow rate and pollutant concentration, and sending the real-time data to the IoT platform.

[0082] Example 2, reference Figure 1 , this embodiment provides a method for treating coal chemical heavy oil wastewater based on the Internet of Things, comprising the following steps:

[0083] Step S1, data collection and monitoring,

[0084] IoT sensors monitor membrane-related real-time data in the sewage treatment system in real time and upload the real-time data to the central control system and IoT platform;

[0085] Real-time data includes inlet and outlet water pressure differential, flux, flow rate, and pollutant concentration;

[0086] Specifically, IoT sensors monitor key parameters in the sewage treatment system in real time, including inlet and outlet water pressure difference, flux, flow rate and pollutant concentration, which are transmitted and analyzed through the central control system and IoT platform to quickly respond to dynamic changes in membrane fouling and reduce the need for human intervention.

[0087] Step S2, membrane fouling trend identification,

[0088] Use big data analysis and machine learning algorithms on the IoT platform to evaluate membrane contamination and predict membrane contamination trends. Membrane contamination includes wastewater composition and pollutant attachment. Based on wastewater composition and pollutant attachment, the current membrane contamination level is evaluated.

[0089] In step S2, the current membrane fouling degree is evaluated:

[0090] Obtain pollutant concentration data from IoT sensors, including:

[0091] Sewage composition: suspended matter concentration C SS , Chemical oxygen demand C COD , Oil pollutant concentration C oil , Salt concentration C salt ;

[0092] Pollutant adhesion: The amount of organic matter attached to the membrane surface A organic 、Inorganic matter A inorganic 、Amount of oily substances attached oil ;

[0093] C SS , reflects the content of particulate matter in sewage, C COD , reflects the concentration of organic matter in sewage, C oil , chemical fouling affecting the membrane, C salt, easy to cause inorganic matter to adhere, A organic , indicating the degree of contamination of the membrane surface by organic pollutants, A inorganic , reflecting the situation of inorganic pollutants blocking the membrane pores, A oil , represents the degree of contamination of the membrane surface by oily substances;

[0094] Extract relevant features of real-time data and calculate the change rate and acceleration of each feature:

[0095] The rate of change, or first-order derivative, is expressed as They respectively represent the changing rates of the data corresponding to the sewage composition and the pollutant attachment conditions;

[0096] The changing acceleration, or second-order derivative, is expressed as They represent the acceleration trends of suspended solids concentration, chemical oxygen demand, the amount of organic matter attached to the membrane surface, and the amount of oil attached;

[0097] In step S2, the current membrane contamination degree is evaluated by:

[0098] Calculate the attachment rate of pollutants: R attachment = k·(C in -C out )·Q, where R attachment is the attachment rate of pollutants, which indicates the accumulation rate of pollutants on the membrane surface per unit time, C in is the pollutant concentration in the influent, C out is the pollutant concentration in the effluent, reflecting the removal effect of the sewage after passing through the membrane, Q is the sewage flow rate, indicating the amount of sewage flowing through the membrane per unit time, and k is the attachment rate coefficient, which is adjusted according to the membrane characteristics and the type of pollutants;

[0099] Based on the attachment rate, the accumulation of membrane pollutants and the degree of membrane surface pollution are calculated:

[0100] Among them, A total (t) is the total amount of pollutants attached to the membrane surface at time t, indicating the accumulation of pollutants on the membrane surface, R attachment (t) is the pollutant attachment rate, t is time, and the integral represents the cumulative effect from the initial moment to the current moment. The total amount of pollutants on the membrane surface is calculated through this formula;

[0101] Specifically, by collecting and analyzing wastewater composition (suspended solids, chemical oxygen demand, oil substances) and pollutant attachment (the amount of organic matter, inorganic matter, and oil substances attached), the current degree of membrane contamination is evaluated, and the calculation of change rate and acceleration is introduced. This makes the evaluation not limited to static analysis, but can also capture the dynamic changes of pollutant concentration and attachment amount.

[0102] In step S2, the membrane fouling trend is predicted as follows:

[0103] The long short-term memory network LSTM is used to capture the long-term and short-term dependencies in the time series and identify the impact of sewage composition and pollutant attachment on membrane fouling. For each moment t, the LSTM model consists of:

[0104] Forget Gate t :f t =σ(W f ·[h t-1 ,x t ]+b f ),

[0105] Input Gate i t :i t =σ(W i ·[h t-1 ,x t ]+b i ),

[0106] Candidate memory cells

[0107] Update memory unit C t :

[0108] Output gate o t :o t =σ(W o ·[h t-1 ,x t ]+b o ), h t =o t tanh(C t );

[0109] Among them, x t Represent input characteristics, including sewage composition and pollutant attachment data characteristics, as well as change rate, change acceleration, attachment rate and accumulation per unit time on the membrane surface;

[0110] h t-1 Indicates the hidden state at the previous moment; W f ,W i ,W C ,W o represents the weight matrix, b f ,b i ,b C ,b o is the bias, σ(·) is the Sigmoid activation function, and tanh(·) is the hyperbolic tangent function;

[0111] In step S2, the method for predicting the membrane fouling trend also includes:

[0112] Through the recursive calculation of the LSTM network, the model finally outputs the predicted value of membrane pollution and predicts the future membrane pollution trend:

[0113] in, is the predicted value of membrane fouling at the future time t+T, indicating the predicted degree of fouling, W o represents the weight matrix of the output layer, h t represents the hidden layer state at the current moment, reflecting the combined influence of pollutant concentration and adhesion, σ(·) is the Sigmoid activation function, b o Represents the bias of the output layer. Through the LSTM model, the membrane pollution trend prediction at the future time t+T is generated to provide a basis for cleaning scheduling and pollution warning;

[0114] Specifically, the long short-term memory network (LSTM) is introduced to predict membrane pollution trends and identify long-term and short-term changes in membrane pollution in advance. The LSTM model can perform deep learning based on historical data and real-time inputs, including sewage composition and pollutant attachment, to predict future membrane pollution trends. It can make cleaning decisions before the pollution situation worsens and avoid a sharp drop in membrane filtration efficiency.

[0115] Step S3, optimization of cleaning cycle and intensity scheduling,

[0116] Based on the current membrane contamination degree generated in step S2, the cleaning cycle and intensity of the membrane separation system are adjusted, the specific time point for starting the cleaning operation is determined, and the amount of cleaning agent used and the flow rate of flushing water are adjusted according to the sewage treatment load and the membrane contamination degree;

[0117] Specifically, based on the pollution assessment and trend prediction results, the cleaning cycle and cleaning intensity can be adjusted dynamically. By intelligently analyzing the current membrane pollution level and sewage treatment load, appropriate cleaning cycles can be selected under different pollution levels.

[0118] The wastewater treatment load here refers to the amount of wastewater and pollutant concentration treated by the membrane in the membrane separation system;

[0119] The cleaning cycle and intensity scheduling method in step S3 is:

[0120] According to the membrane contamination degree calculated in step S2 and the current sewage treatment load, it is determined whether the cleaning operation needs to be started through a preset threshold value;

[0121] According to the degree of contamination, the membrane contamination is divided into mild, moderate and severe, and the corresponding cleaning cycle is set;

[0122] When the membrane fouling level reaches the preset threshold, cleaning is initiated;

[0123] Step S4, dynamic pressure and flow control,

[0124] According to the real-time data in step S1 and the cleaning plan in step S3, the central control system dynamically adjusts the inlet and outlet water pressure and flow rate of the membrane separation system, and automatically adjusts the water pump pressure when the sensor detects a decrease in flux or an increase in pressure difference;

[0125] In step S4, the dynamic pressure and flow rate control method is:

[0126] The sensor monitors the water inlet pressure P in real time in (t), water outlet pressure P out (t), membrane pressure difference ΔP(t) = P in (t)-P out (t), flux J(t) and flow rate Q(t);

[0127] Calculate the pressure difference change rate and flux change rate in the membrane system:

[0128] in, represents the rate of change of pressure difference with time t, C in and C out They represent the pollutant concentrations of the inlet and outlet water, respectively, Q(t) represents the sewage flow rate, which affects the change of the pressure difference, and k1 represents the empirical coefficient, that is, the contribution of the pollutant concentration difference to the pressure difference;

[0129] in, is the rate of change of flux J(t), indicating the change of filtration efficiency, P in (t) is the inlet pressure, which pushes the fluid through the membrane. ΔP(t) is the pressure difference. An increase in the pressure difference reduces the flux. A fouling (t) is the amount of membrane pollutant attachment. As the pollutants attach, the flux gradually decreases. k2 and k3 are empirical coefficients that reflect the effects of inlet pressure and pollutants on flux changes.

[0130] When the pressure difference ΔP(t) increases or the flux J(t) decreases, the pressure and flow of the water pump are automatically adjusted. The water pump pressure regulation formula is:

[0131] Among them, P pump (t) is the pressure of the pump at time t, P pump (t-1) is the pump pressure at the previous moment t-1, is the rate of change of pressure difference. When the pressure difference increases, the pump pressure needs to be increased. is the rate of change of flux. When the flux decreases, the pump pressure also needs to be increased. α and β are adjustment coefficients, which are used to balance the effects of pressure difference and flux change on the pump pressure adjustment.

[0132] Water flow regulation formula: Where, Q(t) is the current water flow, Q set is the set ideal flow rate, ΔP(t) is the real-time pressure difference, ΔP max is the maximum allowable pressure difference. When the pressure difference ΔP(t) is close to ΔP max When the pressure difference increases, the flow rate decreases gradually, and γ is the adjustment coefficient, which controls the degree of flow reduction. When the pressure difference increases, the flow rate is automatically reduced to reduce the load on the membrane, thereby extending the service life of the membrane;

[0133] Specifically, in the case of membrane contamination, sensors are used to monitor pressure difference and flux changes in real time, and automatically adjust the water pump pressure and flow rate when an increase in pressure difference or a decrease in flux is detected. The water pump pressure is automatically adjusted to automatically reduce water flow rate according to the degree of membrane contamination and pressure difference changes, thereby reducing the load on the membrane and further improving the stability and durability of the membrane. The dynamic adjustment mechanism can significantly improve the overall stability and processing capacity of the system under high-load processing conditions, and avoid shutdown failures caused by excessive membrane blockage.

[0134] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A coal chemical heavy oil wastewater treatment system based on the Internet of Things, characterized by: include, The central control system is responsible for receiving data from IoT sensors and membrane contamination assessment results from the IoT platform, issuing operation instructions based on the membrane contamination assessment results, and adjusting the operating status of the membrane separation system, cleaning management unit, and pressure and flow control unit; The central control system includes: The cleaning management unit is responsible for adjusting the amount of cleaning agent, cleaning frequency and flushing water flow rate; The pressure and flow control unit is responsible for adjusting the inlet and outlet water pressure and flow of the membrane separation system; The membrane separation system is responsible for the sewage filtration process. It uses membrane technology to remove suspended solids, oils and macromolecular organic matter in heavy oil sewage. The sewage is driven through the membrane by the inlet and outlet water pressure and flow rate for separation and purification. The IoT platform is responsible for receiving real-time data from IoT sensors, evaluating the degree of membrane contamination based on the real-time data, and transmitting the results of the membrane contamination evaluation to the central control system. IoT sensors are responsible for collecting real-time data in the sewage treatment system, including the membrane inlet and outlet water pressure difference, flux, flow rate and pollutant concentration, and sending the real-time data to the IoT platform.

2. A method for treating coal chemical heavy oil wastewater based on the Internet of Things, characterized in that: A coal chemical heavy oil wastewater treatment system based on the Internet of Things according to claim 1 comprises the following steps: Step S1, data collection and monitoring, the IoT sensor monitors the membrane-related real-time data in the sewage treatment system in real time, and uploads the real-time data to the central control system and the IoT platform; Step S2, membrane pollution trend identification, using big data analysis and machine learning algorithms on the Internet of Things platform to evaluate membrane pollution and predict membrane pollution trends. Membrane pollution includes wastewater composition and pollutant attachment. Based on wastewater composition and pollutant attachment, the current membrane pollution level is evaluated. Step S3, optimizing the cleaning cycle and intensity scheduling, based on the current membrane contamination degree generated in step S2, adjusting the cleaning cycle and intensity of the membrane separation system, determining the specific time point for starting the cleaning operation, and adjusting the amount of cleaning agent used and the flow rate of flushing water according to the sewage treatment load and the membrane contamination degree; Step S4, dynamic pressure and flow control, according to the real-time data in step S1 and the cleaning plan in step S3, the central control system dynamically adjusts the inlet and outlet water pressure and flow of the membrane separation system. When the sensor detects a decrease in flux or an increase in pressure difference, the water pump pressure is automatically adjusted.

3. A method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 2, characterized in that: In step S1, the real-time data includes the inlet and outlet water pressure difference, flux, flow rate and pollutant concentration.

4. A method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 3, characterized in that: In step S2, the current membrane fouling degree is evaluated: Obtain pollutant concentration data from IoT sensors, including: Sewage composition: suspended matter concentration C SS , Chemical oxygen demand C COD , Oil pollutant concentration C oil , Salt concentration C salt ; Pollutant adhesion: The amount of organic matter attached to the membrane surface A organic 、Inorganic matter A inorganic 、Amount of oily substances attached oil ; Extract relevant features of real-time data and calculate the change rate and acceleration of each feature: The rate of change, or first-order derivative, is expressed as They respectively represent the changing rates of the data corresponding to the sewage composition and the pollutant attachment conditions; The changing acceleration, or second-order derivative, is expressed as They respectively represent the acceleration trends of suspended matter concentration, chemical oxygen demand, the amount of organic matter attached to the membrane surface, and the amount of oil attached.

5. A method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 4, characterized in that: In step S2, the current membrane contamination degree is evaluated by: Calculate the attachment rate of pollutants: R attachment = k·(C in -C out )·Q, where R attachment is the attachment rate of pollutants, which indicates the accumulation rate of pollutants on the membrane surface per unit time, C in is the pollutant concentration in the influent, C out is the pollutant concentration in the effluent, reflecting the removal effect of the sewage after passing through the membrane, Q is the sewage flow rate, indicating the amount of sewage flowing through the membrane per unit time, and k is the attachment rate coefficient, which is adjusted according to the membrane characteristics and the type of pollutants; Based on the attachment rate, the accumulation of membrane pollutants and the degree of membrane surface pollution are calculated: Among them, A total (t) is the total amount of pollutants attached to the membrane surface at time t, indicating the accumulation of pollutants on the membrane surface, R attachment (t) is the pollutant attachment rate, t is time, and the integral represents the cumulative effect from the initial moment to the current moment.

6. A method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 5, characterized in that: In step S2, the membrane fouling trend is predicted as follows: The long short-term memory network LSTM is used to capture the long-term and short-term dependencies in the time series and identify the impact of sewage composition and pollutant attachment on membrane fouling. For each moment t, the LSTM model consists of: Forget Gate t :f t =σ(W f ·[h t-1 ,x t ]+b f ), Input Gate i t :i t =σ(W i ·[h t-1 ,x t ]+b i ), Candidate memory cells Update memory unit C t : Output gate o t :o t =σ(W o ·[h t-1 ,x t ]+b o ), h t =o t tanh(C t ); Among them, x t Represent input characteristics, including sewage composition and pollutant attachment data characteristics, as well as change rate, change acceleration, attachment rate and accumulation per unit time on the membrane surface; h t-1 Indicates the hidden state at the previous moment; W f ,W i ,W C ,W o represents the weight matrix, b f ,b i ,b C ,b o is the bias, σ(·) is the Sigmoid activation function, and tanh(·) is the hyperbolic tangent function.

7. A method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 6, characterized in that: In step S2, the method for predicting the membrane fouling trend also includes: Through the recursive calculation of the LSTM network, the model finally outputs the predicted value of membrane pollution and predicts the future membrane pollution trend: in, is the predicted value of membrane contamination at the future time t+T, indicating the predicted degree of contamination, W o represents the weight matrix of the output layer, h t represents the hidden layer state at the current moment, reflecting the combined influence of pollutant concentration and adhesion, σ(·) is the Sigmoid activation function, b o Represents the bias of the output layer, and generates a membrane contamination trend prediction at the future time t+T through the LSTM model.

8. The method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 7, characterized in that: In step S3, the wastewater treatment load refers to the amount of wastewater treated by the membrane in the membrane separation system and the concentration of pollutants.

9. A method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 8, characterized in that: The cleaning cycle and intensity scheduling method in step S3 is: According to the membrane contamination degree calculated in step S2 and the current sewage treatment load, it is determined whether the cleaning operation needs to be started through a preset threshold value; According to the degree of contamination, the membrane contamination is divided into mild, moderate and severe, and the corresponding cleaning cycle is set; When the membrane fouling level reaches a preset threshold, cleaning is initiated.

10. A method for treating coal chemical heavy oil wastewater based on the Internet of Things according to claim 9, characterized in that: In step S4, the dynamic pressure and flow rate control method is: The sensor monitors the water inlet pressure P in real time in (t), water outlet pressure P out (t), membrane pressure difference ΔP(t) = P in (t)-P out (t), flux J(t) and flow rate Q(t); Calculate the pressure difference change rate and flux change rate in the membrane system: in, represents the rate of change of pressure difference with time t, C in and C out They represent the pollutant concentrations of the inlet and outlet water, respectively, Q(t) represents the sewage flow rate, which affects the change of the pressure difference, and k1 represents the empirical coefficient, that is, the contribution of the pollutant concentration difference to the pressure difference; in, is the rate of change of flux J(t), indicating the change of filtration efficiency, P in (t) is the inlet pressure, which pushes the fluid through the membrane. ΔP(t) is the pressure difference. An increase in the pressure difference reduces the flux. A fouling (t) is the amount of membrane pollutant attachment. As the pollutants attach, the flux gradually decreases. k2 and k3 are empirical coefficients that reflect the effects of inlet pressure and pollutants on flux changes. When the pressure difference ΔP(t) increases or the flux J(t) decreases, the pressure and flow of the water pump are automatically adjusted. The water pump pressure regulation formula is: Among them, P pump (t) is the pressure of the pump at time t, P pump (t-1) is the pump pressure at the previous moment t-1, is the rate of change of pressure difference. When the pressure difference increases, the pump pressure needs to be increased. is the rate of change of flux. When the flux decreases, the pump pressure also needs to be increased. α and β are adjustment coefficients, which are used to balance the effects of pressure difference and flux change on the pump pressure adjustment. Water flow regulation formula: Where, Q(t) is the current water flow, Q set is the set ideal flow rate, ΔP(t) is the real-time pressure difference, ΔP max is the maximum allowable pressure difference. When the pressure difference ΔP(t) is close to ΔP max When , the flow rate decreases gradually, and γ is the adjustment coefficient, which controls the degree of flow reduction.

Images