Sewage energy-saving control method and system based on multi-dimensional variable data analysis

Abstract

The invention discloses a sewage energy-saving control method and system based on multi-dimensional variable data analysis, and relates to the technical field of data analysis, and the method comprises the steps: analyzing the incidence relation between variable data in different processing stages, the incidence relation of variables comprises the relation between independent variables and dependent variables and the interaction relation between the independent variables, and the relation between the independent variables and the dependent variables is the relation between the independent variables and the interaction relation between the independent variables; the complex coupling relation among multiple variables is fully and accurately described, and a reliable basis is provided for follow-up regulation and control. And determining the corresponding regulation time delay and variability on each regulation node, and considering the regulation hysteresis and the variable change condition on the sewage treatment flow, thereby setting a threshold value and generating an evaluation index for guiding the subsequent sewage regulation. The regulation and control strategy of each regulation and control node is controlled based on the incidence relation and the regulation and control time delay between the variable data, the adaptability and accuracy of sewage node analysis and control are improved, the real-time sewage treatment effect is ensured, and the sewage treatment efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and particularly to a sewage energy-saving control method and system based on multi-dimensional variable data analysis. Background Art

[0002] The sewage energy-saving control solution based on multi-dimensional variable data analysis relies on modern sensing technology, big data analysis, and machine learning algorithms, aiming to solve the problems of lagging regulation and extensive parameters in traditional sewage treatment. Traditional methods mostly rely on the threshold control of a single index (such as COD, pH), and it is difficult to cope with water quality fluctuations and multi-variable coupling effects, resulting in low treatment efficiency and high energy consumption. The current technology deploys a high-precision sensor network to collect multi-dimensional data such as flow rate, dissolved oxygen, and temperature in real time, and combines time-delay analysis, association rule mining, and other means to reveal the dynamic relationships between variables (such as the time-delay effect between influent water quality and aeration volume). This technology significantly improves the system response speed and anti-interference ability, making sewage treatment shift from experience-driven to data-driven, providing core support for smart water services.

[0003] In the prior art, the threshold control of a single index (such as COD, pH) results in poor adaptability and accuracy of sewage node analysis and control, and cannot guarantee the real-time sewage treatment effect.

[0004] Therefore, how to improve the adaptability and accuracy of sewage node analysis and control is a technical problem to be solved at present. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem of poor adaptability and accuracy of sewage node analysis and control in the prior art, and a sewage energy-saving control method based on multi-dimensional variable data analysis is proposed, which includes: Obtain the sewage treatment process, divide the sewage treatment process into multiple treatment stages, deploy monitoring nodes and regulation nodes on the treatment stages, and determine all variables involved in each treatment stage; Obtain the historical variable data of the sewage treatment process, and analyze the correlation relationship between the variable data of different treatment stages; Analyze the historical variable data to determine the corresponding regulation time-delay and variability on each regulation node, set the variable data threshold of each regulation node through the variability, and generate evaluation indicators; Based on the evaluation indicators, control the regulation strategy of each regulation node based on the correlation relationship between variable data and regulation time-delay, so as to ensure the real-time and efficient sewage treatment.

[0006] In some embodiments of the present application, the treatment stages of the sewage treatment process include a pretreatment stage, a primary treatment stage, a secondary treatment stage, an advanced treatment stage, and a sludge treatment stage, and each treatment stage further includes multiple treatment units.

[0007] In some embodiments of the present application, monitoring nodes and regulation nodes are deployed at the processing stage, including, Determine the processing flow and processing requirements of all processing units at each processing stage, and respectively determine the positions of monitoring nodes and regulation nodes on the processing flow according to the processing flow and processing requirements; Deploy monitoring nodes and regulation nodes respectively through the positions of monitoring nodes and regulation nodes. The monitoring nodes are used to collect and monitor the variable data at the corresponding positions on the processing flow, and the regulation nodes are used to control and adjust the sewage treatment strategy at the corresponding positions on the processing flow.

[0008] In some embodiments of the present application, analyze the correlation relationship between the variable data between different processing stages, including, Distinguish independent variables and dependent variables according to the variable data between different processing stages, establish a matching relationship between independent variables and dependent variables, and describe the correlation relationship between independent variables and dependent variables through a polynomial regression model between independent variables and dependent variables; Divide the model parameters in the polynomial regression model into fixed model parameters and variable model parameters; The fixed model parameters include the polynomial order and the model interaction term; Calculate the non - linear degree between the independent variable and the dependent variable and the noise degree of each independent variable and dependent variable under the matching relationship between each independent variable - dependent variable, determine the overall noise degree of the matching relationship between independent variables - dependent variables according to the noise degree of each independent variable and dependent variable, determine the complexity by combining the non - linear degree and the overall noise degree, and map a polynomial order through the complexity; Determine the combination between independent variables according to the matching relationship between independent variables - dependent variables, detect each independent variable combination through the response surface method, and thus screen out the independent variable combination as the model interaction term; Set the initialization of the variable model parameters according to the independent variable - dependent variable, and train the polynomial regression model based on the initialized variable model parameters and the polynomial order.

[0009] In some embodiments of the present application, set the initialization of the variable model parameters according to the independent variable - dependent variable, and train the polynomial regression model based on the initialized variable model parameters and the polynomial order, including, The variable model parameters include the independent variable coefficient and the interaction term coefficient; Based on the matching relationship between independent variables - dependent variables, calculate the correlation between each independent variable and the dependent variable respectively, calculate the correlation between the independent variables in the model interaction term, denote the correlation between the independent variable and the dependent variable as the first correlation, and denote the correlation between the independent variables in the model interaction term as the second correlation; Determine the initial value of the independent variable coefficient corresponding to each independent variable based on the first correlation, and determine the initial value of the interaction term coefficient corresponding to each model interaction term based on the second correlation; Obtain the independent variable-dependent variable data, and divide the time stamps of the data into a training set and a test set; Define the model structure according to the polynomial order, model interaction terms, initial values of the independent variable coefficients, and initial values of the interaction term coefficients. On the basis of the training set and the test set, continuously optimize the initial values of the independent variable coefficients and the initial values of the interaction term coefficients through the loss function until the loss function converges to obtain the final polynomial regression model.

[0010] In some embodiments of the present application, analyze the historical variable data to determine the corresponding regulation time delay and variability at each regulation node, including, The historical variable data of the regulation node is a historical variable record, and the historical variable record records the variable situation and regulation information that change over time; Mark the multiple regulation positions on the historical variable record according to the regulation information, calculate the cross-correlation function between the regulation information and the variable data, use the time node corresponding to the maximum value of the cross-correlation function as the first time lag value of the single regulation position, combine the multiple regulation positions to obtain the average value of the first time lag value, and use the average value as the first time lag value of the regulation node; Use the multiple regulation positions and the regulation information as a step input, plot the response curve of the variable data, determine the curve time characteristics at the significant change of the variable data on the response curve, and determine the second time lag value of the regulation node according to the curve time characteristics; Combine the first time lag value and the second time lag value of the regulation node to determine the time lag target value; Statistically analyze the variability description parameters of each variable data under the regulation node, integrate all the variability description parameters to determine the variability index of each variable data, and describe the variability of the variable data through the variability index.

[0011] In some embodiments of the present application, set the variable data threshold of each regulation node through variability, including, Determine the basic threshold of each variable data of each regulation node, map a safety margin according to the variability index of each variable data, and superimpose the safety margin on the basic threshold to form a threshold range.

[0012] In some embodiments of the present application, generate evaluation indicators, including, Compare the relationship between the actual value of the variable data and the variable data threshold range to obtain the deviation amount, statistically analyze the regulation response time, and generate an evaluation indicator according to the deviation amount and the regulation response time.

[0013] Correspondingly, the present application further provides a sewage energy-saving control system based on multi-dimensional variable data analysis, including: The first module is used to obtain the sewage treatment process, divide the sewage treatment process into multiple treatment stages, deploy monitoring nodes and regulation nodes on the treatment stages, and determine all variables involved in each treatment stage; The second module is used to obtain the historical variable data of the sewage treatment process and analyze the correlation relationship between the variable data in different treatment stages; The third module is used to analyze the historical variable data to determine the corresponding regulation delay and variability on each regulation node, set the variable data threshold for each regulation node through the variability, and generate evaluation indicators; The fourth module is used to control the regulation strategy of each regulation node based on the correlation relationship between variable data and regulation delay on the basis of the evaluation indicators, so as to ensure the real-time and efficient sewage treatment.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Analyze the correlation relationship between the variable data in different treatment stages. The correlation relationship of the variables here includes the relationship between independent variables and dependent variables and the interaction relationship between independent variables, which fully and accurately describes the complex coupling relationship between multiple variables and provides a reliable basis for subsequent regulation. Determine the corresponding regulation delay and variability on each regulation node, consider the regulation lag on the sewage treatment process and the change of variables, so as to set the threshold and generate evaluation indicators to guide subsequent sewage regulation.

[0015] 2. Control the regulation strategy of each regulation node based on the correlation relationship between variable data and regulation delay, improve the adaptability and accuracy of sewage node analysis and control, ensure the real-time treatment effect of sewage, and improve the sewage treatment efficiency and energy saving. Description of the Drawings

[0016] Figure 1 is a schematic flow chart of the sewage energy-saving control method based on multi-dimensional variable data analysis proposed by the present invention; Figure 2 is a schematic structural diagram of the sewage energy-saving control system based on multi-dimensional variable data analysis proposed by the present invention. Detailed Embodiment

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0018] Refer to Figure 1 , the sewage energy-saving control method based on multi-dimensional variable data analysis includes the following steps: Step S101, obtain the sewage treatment process, divide the sewage treatment process into multiple treatment stages, deploy monitoring nodes and regulation nodes on the treatment stages, and determine all variables involved in each treatment stage.

[0019] In this embodiment, the treatment stage division is as follows: Pretreatment stage: grille, grit chamber (removing large particulate impurities).

[0020] Primary treatment stage: primary sedimentation tank (removing suspended solids, with an SS removal rate of 40 - 50%).

[0021] Secondary treatment stage: bioreactor (activated sludge process or MBR, with a COD removal rate of 85 - 95%).

[0022] Advanced treatment stage: reverse osmosis, advanced oxidation (further removing nitrogen, phosphorus, and micropollutants).

[0023] Sludge disposal stage: sludge thickening, dewatering, incineration (resource utilization).

[0024] In some embodiments of the present application, the treatment stages of the sewage treatment process include a pretreatment stage, a primary treatment stage, a secondary treatment stage, an advanced treatment stage, and a sludge disposal treatment stage, and each treatment stage further includes multiple treatment units.

[0025] In this embodiment, specifically: Purpose: Remove large particulate suspended solids, floating substances, and some organic matters in the sewage, creating favorable conditions for subsequent treatment.

[0026] Main treatment units: Grille: Used to intercept larger suspended solids and floating substances, such as branches, plastics, etc.

[0027] Grit chamber: Remove inorganic sand grains in the sewage through gravitational sedimentation.

[0028] Regulation tank: Regulate the water volume and water quality of the sewage to make the subsequent treatment process more stable.

[0029] Primary sedimentation tank (optional): In some cases, the pretreatment stage may also include a primary sedimentation tank for removing some suspended solids.

[0030] 2. Primary treatment stage Purpose: Further remove suspended solids in the sewage through physical methods, and reduce the BOD (biochemical oxygen demand) and SS (suspended solids) concentrations in the sewage.

[0031] Main treatment units: Primary sedimentation tank (if not set in the pretreatment stage): Remove suspended solids in the sewage through gravitational sedimentation.

[0032] Other physical treatment units: such as dissolved air flotation tanks, etc., which are used to remove grease and fine suspended solids in sewage.

[0033] 3. Secondary treatment stage Purpose: To adopt biological treatment methods to remove organic matters, nutrients such as nitrogen and phosphorus in sewage, and significantly reduce the pollution load of sewage.

[0034] Main treatment units: Activated sludge process: Removing organic matters in sewage through the metabolic action of microorganisms.

[0035] Biofilm process: Such as biological filters, biological rotating discs, etc., using microorganisms on the biofilm to degrade pollutants in sewage.

[0036] Oxidation ditch: An improved activated sludge process with a longer hydraulic retention time and better nitrogen and phosphorus removal effects.

[0037] Sequencing batch reactor (SBR): An activated sludge sewage treatment technology that operates in an intermittent aeration mode.

[0038] 4. Advanced treatment stage Purpose: To further remove refractory organic matters, microorganisms, salts and specific pollutants in sewage, so that the treated water quality meets higher discharge standards or reuse requirements.

[0039] Main treatment units: Filtration: Such as sand filtration, activated carbon filtration, etc., which are used to remove fine suspended solids and dissolved organic matters in sewage.

[0040] Disinfection: Such as ultraviolet disinfection, chlorine disinfection, etc., to kill pathogenic microorganisms in sewage.

[0041] Membrane separation technology: Such as reverse osmosis, ultrafiltration, nanofiltration, etc., removing specific pollutants in sewage through the selective permeability of the membrane.

[0042] Advanced oxidation technology: Such as ozone oxidation, Fenton oxidation, etc., which are used to degrade refractory organic matters in sewage.

[0043] 5. Sludge treatment and disposal stage (accompanying the whole treatment process) Purpose: To treat and dispose of the sludge generated in the sewage treatment process to prevent secondary pollution.

[0044] Main treatment units: Sludge thickening: Reducing the volume of sludge.

[0045] Sludge dewatering: Further reducing the moisture content of sludge to facilitate subsequent disposal.

[0046] Sludge stabilization: Stabilize the organic matter in the sludge through biological or chemical methods.

[0047] Sludge disposal: Dispose of the treated sludge safely, such as landfill, incineration, composting, etc. In some embodiments of the present application, monitoring nodes and control nodes are deployed at the treatment stages, including, Determine the treatment processes and treatment requirements of all treatment units at each treatment stage, and respectively determine the positions of the monitoring nodes and control nodes on the treatment processes according to the treatment processes and treatment requirements; Deploy monitoring nodes and control nodes respectively through the positions of the monitoring nodes and control nodes. The monitoring nodes are used to collect and monitor the variable data at the corresponding positions on the treatment process, and the control nodes are used to control and adjust the sewage treatment strategies at the corresponding positions on the treatment process.

[0048] In this embodiment, monitoring nodes: Deployed at the inlets and outlets of key units (such as the outlet of the primary sedimentation tank, the dissolved oxygen probe of the bioreactor). Control nodes: Deployed at controllable devices (such as aeration blowers, sludge return pumps). Determine the node positions according to the treatment requirements (such as the aeration volume control node is located at the front end of the bioreactor). The monitoring nodes collect data such as flow rate, COD, dissolved oxygen in real time (sampling frequency: 1 time / minute).

[0049] It can be understood that here, for the control node to control and adjust the sewage treatment strategy at the corresponding position on the treatment process, this sewage treatment strategy can be achieved by controlling some chemical dosages or related parameters, such as adjusting the aeration volume, sludge return ratio, chemical dosage, etc.

[0050] Step S102, Obtain the historical variable data of the sewage treatment process, and analyze the correlation relationship between the variable data in different treatment stages.

[0051] In this embodiment, example: influent COD (independent variable) vs dissolved oxygen in secondary treatment (dependent variable). Note that there can be multiple independent variables, and usually one dependent variable involves multiple independent variables. And there may also be an interaction relationship between the independent variables.

[0052] In some embodiments of the present application, analyzing the correlation relationship between the variable data in different treatment stages includes, Distinguish the independent variables and dependent variables according to the variable data in different treatment stages, and establish a matching relationship between the independent variables and the dependent variables. Describe the correlation relationship between the independent variables and the dependent variables through a polynomial regression model between the independent variables and the dependent variables; Divide the model parameters in the polynomial regression model into fixed model parameters and variable model parameters; The fixed model parameters include the polynomial order and the model interaction term; Calculate the non - linear degree between the independent variable and the dependent variable under the matching relationship between each independent variable - dependent variable pair, as well as the noise levels of the independent variable and the dependent variable respectively. Determine the overall noise level of the matching relationship between the independent variable - dependent variable based on the noise levels of the independent variable and the dependent variable respectively. Combine the non - linear degree and the overall noise level to determine the complexity, and map a polynomial order through the complexity. Determine the combinations between independent variables according to the matching relationship between the independent variable - dependent variable. Detect each combination of independent variables through the response surface method, and thus screen out the combination of independent variables as the model interaction term. Set the initialization of the variable model parameters according to the independent variable - dependent variable. Train the polynomial regression model based on the initialized variable model parameters and the polynomial order.

[0053] In this embodiment, the association relationship between the independent variable and the dependent variable is described by a polynomial regression model here. The fixed model parameters are the parameters that need to be determined in advance, and the variable model parameters are the parameters that are continuously optimized through subsequent data training. The non - linear degree can be described by the cross - correlation function or the partial correlation coefficient, etc. Combine the non - linear degree and the overall noise level to determine the complexity (after normalization, calculate by means of weighted summation, etc.). If the data shows an obvious non - linear relationship, a higher polynomial order is required to capture this non - linearity. Data with a large amount of noise may require a lower polynomial order to avoid overfitting the noise.

[0054] In this embodiment, there may be an interaction (mutual influence) between independent variables. For the response surface method (RSM), draw a three - dimensional response surface diagram and observe the degree of curvature of the surface (such as a saddle point indicating a strong interaction).

[0055] In some embodiments of the present application, set the initialization of the variable model parameters according to the independent variable - dependent variable, and train the polynomial regression model based on the initialized variable model parameters and the polynomial order, including, The variable model parameters include the independent variable coefficient and the interaction term coefficient; Based on the matching relationship between the independent variable - dependent variable, calculate the correlation between each independent variable and the dependent variable respectively, and calculate the correlation between the independent variables in the model interaction term. Denote the correlation between the independent variable and the dependent variable as the first correlation, and denote the correlation between the independent variables in the model interaction term as the second correlation; Determine the initial value of the independent variable coefficient corresponding to each independent variable based on the first correlation, and determine the initial value of the interaction term coefficient corresponding to each model interaction term based on the second correlation; Obtain the data of the independent variable - dependent variable, and divide the time stamps of its data into a training set and a test set; Define the model structure according to the polynomial order, model interaction terms, initial values of independent variable coefficients, and initial values of interaction term coefficients. Based on the training set and test set, continuously optimize the initial values of independent variable coefficients and initial values of interaction term coefficients through the loss function until the loss function converges to obtain the final polynomial regression model.

[0056] In this embodiment, the correlation can be described by the Pearson correlation coefficient. The training steps are as follows: 1. Calculate the correlation between the independent variable and the dependent variable (such as the Pearson coefficient).

[0057] 2. Initialize the coefficients (high correlation corresponds to high initial values).

[0058] 3. Divide the training set (70%) and the test set (30%), and use the mean squared error (MSE) as the loss function.

[0059] 4. Iteratively optimize the coefficients until the MSE converges.

[0060] For example, for the relationship between a certain two independent variables and the dependent variable, the polynomial regression model is as follows: ; Wherein, is the dependent variable, is the intercept term, , are respectively , the independent variable coefficients of is the interaction term coefficient, , are respectively the two independent variables, is the error term. It should be noted that the polynomial regression model can be adjusted according to the specific variable data relationship and can be multiple times. The intercept term and the error term are continuously optimized through training.

[0061] Step S103: Analyze the historical variable data to determine the corresponding regulation delay and variability at each regulation node, and set the variable data threshold at each regulation node through the variability to generate evaluation indicators.

[0062] In this embodiment, due to the continuity and complexity of the sewage treatment process, there will be regulation delay, that is, the effect of a certain regulation will be reflected after a period of time at the regulation point and be captured by the data. Cross-correlation function method: Calculate the cross-correlation function between the regulation action and the variable response, and the peak corresponds to the time lag. Step response method: Plot the variable response curve after regulation to determine the significant change time point. Variability refers to the change situation of the variable data. Because of the fluctuation of the sewage treatment process, a fixed variable data threshold is difficult to accurately describe the rationality of sewage treatment regulation. Consider the change situation of the variable data to set the variable data threshold.

[0063] In some embodiments of the present application, analyzing historical variable data to determine the corresponding regulation time delay and variability at each regulation node, including: The historical variable data of the regulation node is a historical variable record, on which the variable situation changing with time and the regulation information are recorded; Mark multiple regulation positions on the historical variable record according to the regulation information, calculate the cross-correlation function between the regulation information and the variable data, take the time node corresponding to the maximum value of the cross-correlation function as the first time lag value of the single regulation position, combine the multiple regulation positions to obtain the average value of the first time lag value, and take the average value as the first time lag value of the regulation node; Take the multiple regulation positions and the regulation information as a step input, plot the response curve of the variable data, determine the curve time characteristics at the significant change of the variable data on the response curve, and determine the second time lag value of the regulation node according to the curve time characteristics; Combine the first time lag value and the second time lag value of the regulation node to determine the time lag target value; Statistically analyze the variability description parameters of each variable data under the regulation node, integrate all the variability description parameters to determine the variability index of each variable data, and describe the variability of the variable data through the variability index.

[0064] In this embodiment, the cross-correlation function is used to measure the correlation between two signals at different time lags. Calculate the cross-correlation function between the regulation measure signal and the variable data signal to obtain a series of correlation coefficients at different time lags. Find the maximum value in the cross-correlation function, and the time lag corresponding to this value is the initially estimated time lag. It should be noted that there may be multiple peaks in the cross-correlation function, and it is necessary to judge which peak is the real lag time in combination with the actual situation. Regard the regulation measure as a step input and observe the response curve of the variable data. The response curve usually includes characteristics such as rise time, peak time, and steady time. Measure the time point when the variable data starts to change significantly from the start of the measure implementation. The significant change can be defined as the data change exceeding a certain threshold or the change rate exceeding a certain value. Determine the second time lag value of the regulation node according to all the curve time characteristics, and combine the first time lag value and the second time lag value of the regulation node to determine the time lag target value. The specific calculation formula is as follows: ; Wherein, is the time lag target value, , are the lag weights of the first time lag value and the second time lag value respectively, , are the first time lag value and the second time lag value respectively, is , The larger value of the two, is a preset constant, indicating the correction of the larger value to the average of the two, which is to balance the magnitude of the correction function and can more accurately combine the two methods to determine the latency.

[0065] In this embodiment, the variability description parameters include three categories: amplitude, frequency, and trend. Amplitude analysis: Calculate the change amount: Calculate the change amount of the variable data after the implementation of the control measures, such as the difference between the maximum value and the minimum value, the change of the average value, etc.

[0066] Statistical distribution: Analyze the statistical distribution of the change amount, such as the mean value, standard deviation, quantile, etc., to understand the overall situation of the change.

[0067] Frequency analysis: Number of fluctuations: Count the number of fluctuations of the variable data within a unit time to understand the frequency of the change.

[0068] Periodicity analysis: Use methods such as Fourier transform and wavelet analysis to extract the periodic components of the variable data to identify whether there are periodic changes.

[0069] Trend analysis: Long-term trend: Analyze the change trend of the variable data within a long time range, such as rising, falling, or stable.

[0070] Short-term fluctuation: Pay attention to the fluctuation of the variable data within a short time to understand the immediate impact of the control measures on the data.

[0071] Determine a change index by integrating the three categories of amplitude, frequency, and trend.

[0072] In some embodiments of the present application, the variable data threshold of each control node is set by variability, including, Determine the basic threshold of each variable data of each control node, map a safety margin according to the change index of each variable data, and superimpose the safety margin on the basic threshold to form a threshold range.

[0073] In this embodiment, the basic threshold of the variable data can be determined by historical data, process requirement safety redundancy, etc. The larger the change index, the larger the safety margin. Superimpose the safety margin on the basic threshold to adjust the upper and lower limits of the basic threshold to form a threshold range.

[0074] In some embodiments of the present application, evaluation indicators are generated, including, Compare the actual value of the variable data with the threshold range of the variable data to obtain the deviation amount, count the control response time, and generate an evaluation index based on the deviation amount and the control response time.

[0075] In this embodiment, the deviation amount: the degree of deviation between the actual value and the threshold range (for example, for dissolved oxygen of 2.6 mg / L, the deviation is +0.1 mg / L). The control response time: the time from the control action to the variable entering the threshold range (for example, the dissolved oxygen stabilizes 12 minutes after the aeration volume is adjusted). An evaluation index is generated based on the deviation amount and the control response time of each variable data (select the variable data that can be used for evaluation from all variable data in this link). The specific formula is as follows: ; Wherein, is the evaluation index of the th control node, is the number of variable data that can be used for evaluation of the th control node, is the evaluation weight of the th variable data, is the th th deviation amount of the th control node, is the th control response time of the th th variable data of the th control node,

[0076]

[0077] In this embodiment, by analyzing the current treatment effect through the evaluation index, it can be specifically a certain variable data. The control strategy can be adjusted by the proportional-integral-differential control (PID) method. The aeration volume is dynamically adjusted according to the DO deviation: quickly adjusted when the deviation is large (proportional P), cumulatively adjusted when the deviation persists (integral I), and pre-judgment adjusted when the deviation change rate is large (differential D). Example: If the DO is lower than the set value by 0.5 mg / L, the PID output increases the aeration volume by 10%. For fuzzy control, the variable is fuzzified into "high / medium / low", and an instruction is generated based on the rule base: Rule: If the DO is low and the COD is high, then increase the aeration volume.

[0078] Correspondingly, the present application further provides a sewage energy-saving control system based on multi-dimensional variable data analysis, as Figure 2 shown, including A first module for obtaining the sewage treatment process, dividing the sewage treatment process into multiple treatment stages, deploying monitoring nodes and regulation nodes on the treatment stages, and determining all variables involved in each treatment stage; A second module for obtaining historical variable data of the sewage treatment process and analyzing the correlation relationship between the variable data in different treatment stages; A third module for analyzing the historical variable data to determine the corresponding regulation delay and variability on each regulation node, setting the variable data threshold for each regulation node through the variability, and generating evaluation indicators; A fourth module for controlling the regulation strategy of each regulation node based on the evaluation indicators, the correlation relationship between the variable data, and the regulation delay, so as to ensure the real-time and efficient sewage treatment.

[0079] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Analyze the correlation relationship between the variable data in different treatment stages. The correlation relationship of the variables here includes the relationship between the independent variable and the dependent variable and the interaction relationship between the independent variables, which fully and accurately describes the complex coupling relationship between multiple variables and provides a reliable basis for subsequent regulation. Determine the corresponding regulation delay and variability on each regulation node, consider the regulation lag on the sewage treatment process and the change of variables, and thus set the threshold to generate evaluation indicators to guide subsequent sewage regulation.

[0080] 2. Control the regulation strategy of each regulation node based on the correlation relationship between the variable data and the regulation delay, improve the adaptability and accuracy of sewage node analysis and control, ensure the real-time treatment effect of sewage, and improve the sewage treatment efficiency.

[0081] Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present invention.

[0082] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present invention.

[0083] Those skilled in the art can understand that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed to be located in one or more systems different from this implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0084] As mentioned above, the above are only the preferred specific implementation manners of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent replacements or changes should be covered within the protection scope of the present invention.

Claims

1. A sewage energy-saving control method based on multi-dimensional variable data analysis, characterized in that, including Obtain the sewage treatment process, divide the sewage treatment process into multiple treatment stages, deploy monitoring nodes and regulation nodes on the treatment stages, and determine all variables involved in each treatment stage; Obtain the historical variable data of the sewage treatment process, and analyze the correlation relationship between the variable data of different treatment stages; Analyze the historical variable data to determine the corresponding regulation delay and variability on each regulation node, set the variable data threshold of each regulation node through the variability, and generate evaluation indicators; Based on the evaluation indicators, control the regulation strategy of each regulation node based on the correlation relationship and regulation delay between the variable data, so as to ensure the real-time and efficient sewage treatment.

2. The sewage energy-saving control method based on multi-dimensional variable data analysis according to claim 1, characterized in that The treatment stages of the sewage treatment process include the pretreatment stage, the primary treatment stage, the secondary treatment stage, the advanced treatment stage and the sludge disposal treatment stage, and each treatment stage also includes multiple treatment units.

3. The sewage energy-saving control method based on multi-dimensional variable data analysis according to claim 2, characterized in that Deploy monitoring nodes and regulation nodes on the treatment stages, including Determine the treatment process and treatment requirements of all treatment units in each treatment stage, and respectively determine the positions of monitoring nodes and regulation nodes on the treatment process according to the treatment process and treatment requirements; Deploy monitoring nodes and regulation nodes respectively through the positions of monitoring nodes and regulation nodes. The monitoring nodes are used to collect and monitor the variable data at the corresponding positions on the treatment process, and the regulation nodes are used to control and adjust the sewage treatment strategy at the corresponding positions on the treatment process.

4. The sewage energy-saving control method based on multi-dimensional variable data analysis according to claim 1, characterized in that Analyze the correlation relationship between the variable data of different treatment stages, including Distinguish independent variables and dependent variables according to the variable data between different treatment stages, establish a matching relationship between independent variables and dependent variables, and describe the correlation relationship between independent variables and dependent variables through a polynomial regression model between independent variables and dependent variables; Divide the model parameters in the polynomial regression model into fixed model parameters and variable model parameters; The fixed model parameters include the polynomial order and the model interaction term; Calculate the non-linear degree between the independent variable and the dependent variable and the noise degree of each of the independent variable and the dependent variable under the matching relationship between each independent variable and the dependent variable, determine the overall noise degree of the matching relationship between the independent variable and the dependent variable according to the noise degree of each of the independent variable and the dependent variable, combine the non-linear degree and the overall noise degree to determine the complexity, and map a polynomial order through the complexity; Determine the combination between independent variables according to the matching relationship between independent variables and dependent variables, detect each independent variable combination through the response surface method, and thus screen out the independent variable combination as the model interaction term; Set the initialization of the variable model parameters according to the independent variable and the dependent variable, and train the polynomial regression model based on the initialized variable model parameters and the polynomial order.

5. The sewage energy-saving control method based on multi-dimensional variable data analysis according to claim 4, characterized in that Set the initialization of the variable model parameters according to the independent variable and the dependent variable, and train the polynomial regression model based on the initialized variable model parameters and the polynomial order, including The variable model parameters include the independent variable coefficient and the interaction term coefficient; Based on the matching relationship between the independent variables and the dependent variable, calculate the correlation between each independent variable and the dependent variable respectively, and calculate the correlation between the independent variables in the model interaction term. Denote the correlation between the independent variable and the dependent variable as the first correlation, and denote the correlation between the independent variables in the model interaction term as the second correlation; Determine the initial value of the independent variable coefficient corresponding to each independent variable based on the first correlation, and determine the initial value of the interaction term coefficient corresponding to each model interaction term based on the second correlation; Obtain the data of the independent variable - dependent variable, and divide the time stamps of the data into a training set and a test set; Define the model structure according to the polynomial order, model interaction terms, initial values of the independent variable coefficients, and initial values of the interaction term coefficients. Based on the training set and the test set, continuously optimize the initial values of the independent variable coefficients and the initial values of the interaction term coefficients through the loss function until the loss function converges to obtain the final polynomial regression model.

6. The sewage energy-saving control method based on multi-dimensional variable data analysis according to claim 1, wherein Analyze the historical variable data to determine the corresponding regulation time delay and variability at each regulation node, including, The historical variable data of the regulation node is a historical variable record, and the historical variable record records the variable situation and regulation information that change over time; Mark the multiple regulation positions on the historical variable record according to the regulation information, calculate the cross - correlation function between the regulation information and the variable data, take the time node corresponding to the maximum value of the cross - correlation function as the first time lag value of the single regulation position, and combine the multiple regulation positions to obtain the average value of the first time lag value, and take the average value as the first time lag value of the regulation node; Take the multiple regulation positions and the regulation information as a step input, and draw the response curve of the variable data. Determine the curve time characteristics at the significant change of the variable data on the response curve, and determine the second time lag value of the regulation node according to the curve time characteristics; Combine the first time lag value and the second time lag value of the regulation node to determine the time lag target value; Statistically analyze the variability description parameters of each variable data under the regulation node, integrate all the variability description parameters to determine the variability index of each variable data, and describe the variability of the variable data through the variability index.

7. The sewage energy-saving control method based on multi-dimensional variable data analysis according to claim 6, characterized in that Set the variable data threshold for each regulation node through variability, including, Determine the basic threshold of each variable data for each regulation node, map a safety margin according to the variability index of each variable data, and add the safety margin to the basic threshold to form a threshold range.

8. The sewage energy-saving control method based on multi-dimensional variable data analysis according to claim 7, characterized in that, Generate evaluation indicators, including, Compare the relationship between the actual value of the variable data and the variable data threshold range to obtain the deviation amount, statistically analyze the regulation response time, and generate evaluation indicators according to the deviation amount and the regulation response time.

9. The sewage energy-saving control system based on multi-dimensional variable data analysis is characterized in that including, The first module is used to obtain the sewage treatment process, divide the sewage treatment process into multiple treatment stages, deploy monitoring nodes and regulation nodes on the treatment stages, and determine all variables involved in each treatment stage; The second module is used to obtain the historical variable data of the sewage treatment process and analyze the correlation relationship between the variable data in different treatment stages; The third module is used to analyze historical variable data to determine the corresponding regulation delay and variability on each regulation node, set the variable data threshold for each regulation node through the variability, and generate evaluation indicators; The fourth module is used to control the regulation strategy of each regulation node based on the correlation between variable data and regulation delay on the basis of the evaluation indicators, so as to ensure the real-time and efficient sewage treatment.

Images