Phosphate intelligent addition method and system for low phosphorus boiler water

Abstract

The present application relates to the technical field of intelligent control, and particularly relates to a phosphate intelligent adding method and system for low-phosphorus boiler water. The conductivity, pH and phosphate parameters in the boiler water are collected, and the use data sequence of the boiler is collected; according to the use data sequence, the corrosion of the boiler is predicted to obtain the predicted corrosion parameters; the conductivity correction coefficient and the phosphate correction coefficient are generated according to the predicted corrosion parameters, the conductivity and the phosphate parameters are corrected to obtain the corrected conductivity and the corrected phosphate parameters; the latest use data is extracted from the use data sequence to predict the characteristics of the boiler water, the predicted conductivity and the predicted phosphate parameters are obtained, the similarity of the corrected conductivity and the corrected phosphate parameters is analyzed, the parameter weight and the corrosion weight are configured, the phosphate addition is optimized in combination with the predicted corrosion parameters, the optimized phosphate addition parameters are obtained, and the phosphate is added. The intelligent and accurate addition of the phosphate for the low-phosphorus boiler water is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, and particularly relates to a phosphate intelligent adding method and system for low-phosphorus boiler water. BACKGROUND

[0002] In industrial production, as an important heat energy equipment, the water quality treatment of boiler water is very important. In order to prevent the problems such as boiler fouling and corrosion, it is usually necessary to add phosphate and other medicaments to the boiler water. The traditional phosphate adding mode depends on manual experience or simple automatic control mode, and the adding operation is performed according to some basic boiler water parameters, so it is difficult to accurately add phosphate according to the actual state of the boiler, which may lead to insufficient addition and inability to effectively prevent boiler corrosion, or excessive addition, causing waste of medicaments and deterioration of boiler water quality. There are technical problems of insufficient accuracy of measurement and utilization of boiler water parameters, poor timeliness and adaptability of phosphate addition control. SUMMARY

[0003] The present application provides a phosphate intelligent adding method and system for low-phosphorus boiler water to solve the technical problems of insufficient accuracy of measurement and utilization of boiler water parameters, poor timeliness and adaptability of phosphate addition in the prior art.

[0004] The technical scheme for solving the above technical problems of the present application is as follows:

[0005] In a first aspect, the present application provides a phosphate intelligent adding method for low-phosphorus boiler water, comprising: testing and collecting the conductivity, pH and phosphate parameters in the boiler water, and collecting a usage data sequence of the boiler; performing boiler corrosion prediction according to the usage data sequence to obtain a predicted corrosion parameter; generating a conductivity correction coefficient and a phosphate correction coefficient according to the predicted corrosion parameter, correcting the conductivity and phosphate parameters to obtain corrected conductivity and corrected phosphate parameters; extracting the latest usage data in the usage data sequence to perform boiler water feature prediction, obtaining predicted conductivity and predicted phosphate parameters, analyzing the similarity with the corrected conductivity and corrected phosphate parameters, configuring a parameter weight and a corrosion weight, combining the predicted corrosion parameter to perform phosphate addition optimization to obtain optimized phosphate addition parameters, and performing phosphate addition.

[0006] Optionally, the testing and collecting of the conductivity, pH and phosphate parameters in the boiler water, and the collecting of the usage data sequence of the boiler, comprise: testing and collecting the conductivity, pH and phosphate parameters in the boiler water; and obtaining the usage data sequence after the boiler is put into use, wherein each usage data comprises boiler operation data in a usage time period.

[0007] The boiler corrosion prediction is performed according to the use data sequence, and a predicted corrosion parameter is obtained, including: obtaining a boiler aging predictor; inputting the use data sequence into the boiler aging predictor, and outputting the predicted corrosion parameter obtained in the boiler aging predictor.

[0008] The obtaining step of the boiler aging predictor includes: collecting a sample use data sequence set according to boiler operation data in a historical time, and labeling corrosion sizes of the boiler under different sample use data sequences to obtain a sample corrosion parameter set; taking the use data sequence as input data and the corrosion parameter as output data to construct a boiler aging predictor based on machine learning; and supervising training of the boiler aging predictor by using the sample use data sequence set and the sample corrosion parameter set until the accuracy converges.

[0009] Optionally, the conductivity correction coefficient and the phosphate correction coefficient are generated according to the predicted corrosion parameter, and the conductivity and the phosphate parameter are corrected to obtain corrected conductivity and corrected phosphate parameters, including: inputting the predicted corrosion parameter into a parameter correction classification table to output the conductivity correction coefficient and the phosphate correction coefficient, wherein the parameter correction classification table includes a conductivity correction classification sub-table and a phosphate correction classification sub-table, and is constructed based on a mapping relationship between a sample corrosion parameter set and sample conductivity error coefficient set and sample phosphate error coefficient set in boiler operation historical data; the predicted corrosion parameter is input into the parameter correction classification table to obtain the conductivity correction coefficient and the phosphate correction coefficient by mapping classification; and the conductivity correction coefficient and the phosphate correction coefficient are used to correct and calculate the conductivity and the phosphate parameter respectively to obtain the corrected conductivity and the corrected phosphate parameter.

[0010] Optionally, the latest use data is extracted from the use data sequence, the boiler water feature prediction is performed, the predicted conductivity and the predicted phosphate parameter are obtained, the similarity with the corrected conductivity and the corrected phosphate parameter is analyzed, and the parameter weight and the corrosion weight are configured, including: extracting the latest use data from the use data sequence; inputting the latest use data into a boiler water feature predictor to output the predicted conductivity and the predicted phosphate parameter, wherein the boiler water feature predictor is trained by using a sample use data set, a sample predicted conductivity set and a sample predicted phosphate parameter set; calculating the average similarity of the corrected conductivity and the corrected phosphate parameter with the predicted conductivity and the predicted phosphate parameter; and taking the average similarity as the parameter weight and calculating the corrosion weight.

[0011] Optionally, in combination with the predicted corrosion parameter, the phosphate addition optimization is performed to obtain an optimized phosphate addition parameter, and the phosphate addition is performed, including: randomly generating a first phosphate addition parameter; according to the predicted conductivity, the predicted phosphate parameter, the corrected conductivity and the corrected phosphate parameter, a fused conductivity and a fused phosphate parameter are calculated and obtained; according to the pH, the fused conductivity and the fused phosphate parameter, a basic phosphate concentration is processed and obtained, in combination with the first phosphate addition parameter, a first added phosphate concentration is calculated and obtained, and a similarity with a standard phosphate concentration is calculated to obtain a first addition fitness; according to the first added phosphate concentration in combination with the predicted corrosion parameter, a first corrosion protection coefficient is obtained by corrosion protection prediction; according to the parameter weight and the corrosion weight, the first addition fitness and the first corrosion protection coefficient are weighted and calculated to obtain a first phosphate fitness; the iteration optimization of the phosphate addition parameter is continued until convergence, and the optimized phosphate addition parameter with the maximum phosphate fitness is obtained to perform the phosphate addition.

[0012] According to the first added phosphate concentration in combination with the predicted corrosion parameter, a first corrosion protection coefficient is obtained by corrosion protection prediction, including: according to the boiler phosphate addition data in the historical time, a sample corrosion parameter set and a sample phosphate concentration set are collected, and the size ratio of the protective layer in the boiler is tested and obtained under different sample corrosion parameters and sample phosphate concentrations, and a sample corrosion protection coefficient set is labeled; a machine learning is used to construct a corrosion protection predictor, and supervised training is performed based on the sample corrosion parameter set, the sample phosphate concentration set and the sample corrosion protection coefficient set until the requirements are met; the first added phosphate concentration and the predicted corrosion parameter are input into the corrosion protection predictor, and the first corrosion protection coefficient is output.

[0013] In a second aspect, the present application provides a phosphate intelligent addition system for low-phosphorus boiler water, including:

[0014] A key data acquisition module is used to test and collect the conductivity, pH, phosphate parameter in the boiler water, and collect the use data sequence of the boiler;

[0015] A predicted corrosion parameter acquisition module is used to perform boiler corrosion prediction according to the use data sequence to obtain a predicted corrosion parameter;

[0016] A measurement parameter correction module is used to generate a conductivity correction coefficient and a phosphate correction coefficient according to the predicted corrosion parameter, and correct the conductivity and the phosphate parameter to obtain a corrected conductivity and a corrected phosphate parameter;

[0017] The phosphate addition optimization module is configured to extract latest use data from the use data sequence, perform a boiler water feature prediction, obtain a predicted conductivity and a predicted phosphate parameter, analyze similarity with the corrected conductivity and the corrected phosphate parameter, configure a parameter weight and a corrosion weight, perform phosphate addition optimization in combination with the predicted corrosion parameter, obtain an optimized phosphate addition parameter, and perform phosphate addition.

[0018] By implementing the present application, the conductivity, pH, and phosphate parameter in the boiler water can be tested and collected, and the use data sequence of the boiler can be collected, ensuring that subsequent analysis and calculation have accurate and comprehensive data support, laying a data foundation for the entire phosphate intelligent addition process, and guaranteeing the reliability of subsequent steps.

[0019] By implementing the present application, the boiler corrosion prediction can be performed according to the use data sequence, the predicted corrosion parameter can be obtained, the possible corrosion condition of the boiler can be known in advance, the subsequent parameter correction and phosphate addition are more targeted, the corrosion problem can be better addressed, and the service life of the boiler is prolonged.

[0020] By implementing the present application, the conductivity correction coefficient and the phosphate correction coefficient can be generated according to the predicted corrosion parameter, the conductivity and the phosphate parameter can be corrected, the corrected conductivity and the corrected phosphate parameter can be obtained, the parameter measurement error caused by corrosion and other factors can be eliminated, the conductivity and the phosphate parameter are more accurate, and accurate data basis is provided for subsequent phosphate addition optimization.

[0021] By implementing the present application, the latest use data can be extracted from the use data sequence, the boiler water feature prediction can be performed, the predicted conductivity and the predicted phosphate parameter can be obtained, the similarity with the corrected conductivity and the corrected phosphate parameter can be analyzed, the parameter weight and the corrosion weight can be configured, the phosphate addition optimization can be performed in combination with the predicted corrosion parameter, the optimized phosphate addition parameter can be obtained, and the phosphate addition can be performed, so that the amount of phosphate added is more accurate, the treatment needs of the boiler water can be met, problems caused by excessive or insufficient addition can be avoided, and the corrosion protection effect on the boiler is enhanced in combination with the predicted corrosion parameter.

[0022] In summary, by implementing the present application, intelligent and accurate addition of phosphate for low-phosphorus boiler water can be realized, the corrosion protection capability of the boiler is effectively improved, and the stable operation of the boiler is guaranteed. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 A flowchart of a phosphate intelligent addition method for low-phosphorus boiler water provided by the present application is shown.

[0024] Figure 2 A structural diagram of a phosphate intelligent addition system for low-phosphorus boiler water provided by the present application is shown.

[0025] In the drawings, the components represented by the respective reference numerals are as follows:

[0026] Key data acquisition module 11, predicted corrosion parameter acquisition module 12, measured parameter correction module 13, phosphate addition optimization module 14. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0028] In the description of the present application, the terms "first", "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" can explicitly or implicitly include one or more features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0029] In the description of the present application, the term "for example" is used to indicate "as an example, illustration or explanation". Any embodiment described as "for example" in the present application is not necessarily interpreted as more preferred or more advantageous than other embodiments. The following description is given in order to enable any person skilled in the art to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that those skilled in the art can realize the present application without using these specific details. In other examples, well-known structures and processes will not be described in detail in order to avoid unnecessary details making the description of the present application obscure. Therefore, the present application is not intended to be limited to the shown embodiments, but is consistent with the broadest scope of principles and characteristics disclosed.

[0030] Embodiment one, as shown in the present application, provides a phosphate intelligent addition method for low-phosphorus boiler water, comprising: Figure 1

[0031] S100: Test and collect the conductivity, pH, and phosphate parameters in the boiler water, and collect the usage data sequence of the boiler;

[0032] S200: According to the usage data sequence, perform boiler corrosion prediction to obtain predicted corrosion parameters;

[0033] ​S300: generating conductivity correction coefficients and phosphate correction coefficients according to the predicted corrosion parameters, correcting the conductivity and phosphate parameters, and obtaining corrected conductivity and corrected phosphate parameters;

[0034] S400: extracting the latest use data in the use data sequence, performing a boiler water feature prediction to obtain predicted conductivity and predicted phosphate parameters, analyzing the similarity with the corrected conductivity and corrected phosphate parameters, configuring parameter weights and corrosion weights, combining the predicted corrosion parameters to optimize phosphate addition, obtaining optimized phosphate addition parameters, and performing phosphate addition.

[0035] In step S100 of the embodiments of the present application, the conductivity, pH, and phosphate parameters in the boiler water are tested and collected, and the use data sequence of the boiler is collected, including:

[0036] The conductivity, pH, and phosphate parameters in the boiler water are tested and collected.

[0037] The use data sequence after the boiler is put into use is obtained, wherein each use data includes the boiler operation data in a use time period.

[0038] In the embodiments of the present application, the purpose of step S100 is to collect the conductivity, pH, and phosphate parameters in the boiler water and collect the use data sequence of the boiler, determine whether the boiler water is in a reasonable operation range, provide historical basis for subsequent analysis of the operation law, aging trend, and corrosion risk of the boiler, and ensure that the subsequent boiler corrosion prediction, phosphate parameter correction, and phosphate addition optimization can be combined with the actual use of the boiler, rather than relying only on the current instantaneous data.

[0039] The conductivity, pH, and phosphate parameters in the boiler water can be directly obtained by using the conductivity sensor, pH sensor, and phosphate online analyzer installed in the boiler water circulation system to test the boiler water sample at regular intervals. For the phosphate parameter, the spectrophotometric method is used for determination, that is, by measuring the absorption degree of phosphate ions to light at a specific wavelength to determine the concentration of phosphate in the boiler water.

[0040] The use data sequence after the boiler is put into use can be collected by the data recording system in the boiler operation process, and each use data includes the boiler operation data in a use time period, such as operation time, load size, furnace temperature, and pipeline pressure.

[0041] For example, during the operation of a boiler in a certain factory, the conductivity of the boiler water is measured every certain time, such as every 1 hour, using a conductivity meter, the pH value of the boiler water is measured using a pH meter, and the phosphate radical concentration of the boiler water is measured using a spectrophotometric phosphate radical online analyzer. At the same time, the data recording system of the boiler records the operation time, load size, furnace temperature, pipe pressure and other use data of the boiler in real time every hour, forming a use data sequence of the boiler. These data will be used for subsequent boiler corrosion prediction and phosphate addition optimization and the like.

[0042] In step S200 of the embodiment of the present application, according to the use data sequence, the boiler corrosion prediction is performed to obtain a predicted corrosion parameter, including:

[0043] obtaining a boiler aging predictor;

[0044] inputting the use data sequence into the boiler aging predictor, inputting into the boiler aging predictor, and outputting to obtain a predicted corrosion parameter.

[0045] In step S200 of the embodiment of the present application, the obtaining step of the boiler aging predictor includes:

[0046] According to the boiler operation data in the historical time, a sample use data sequence set is collected, and the corrosion size of the boiler under different sample use data sequences is labeled to obtain a sample corrosion parameter set;

[0047] using the use data sequence as input data and the corrosion parameter as output data, a boiler aging predictor based on machine learning is constructed;

[0048] The sample use data sequence set and the sample corrosion parameter set are used to supervise the training of the boiler aging predictor until the accuracy converges.

[0049] In the embodiment of the present application, the boiler aging predictor is obtained, which provides a reliable model tool for subsequent boiler corrosion prediction based on the use data sequence, ensures the accuracy of the predicted corrosion parameter, and further lays a foundation for subsequent parameter correction and phosphate addition optimization.

[0050] First, sample data for training the boiler aging predictor needs to be collected, that is, according to the boiler operation data in the historical time, a sample use data sequence set is collected, and the corrosion size of the boiler under different sample use data sequences is labeled to obtain a sample corrosion parameter set. The sample use data sequence set and the sample corrosion parameter set are used to train the boiler aging predictor.

[0051] Specifically, historical boiler operation data in a historical time period needs to be collected to form a sample usage data sequence set, and each sample usage data sequence contains boiler operation data in a corresponding time period, such as the above-mentioned operation time, load size, furnace temperature, pipeline pressure and other parameters.

[0052] For example, assuming that a power plant has 10 boilers of the same type and the operation history is 5 years, the monthly operation data of the 10 boilers of the same type in the past 5 years, such as monthly cumulative operation time, average load, maximum temperature, average pipeline pressure, etc., can be collected to form a sample usage data sequence set of 10 groups x 60 months.

[0053] Then, for each sample usage data sequence, the maximum corrosion depth, corrosion area and other corrosion sizes of the corresponding boiler are obtained through actual detection means such as non-destructive testing and corrosion site measurement, and are labeled as sample corrosion parameters to form a sample corrosion parameter set. For example, through annual shutdown detection, the maximum corrosion depth of each boiler each year is recorded, such as 0.1 mm in the first year and 0.15 mm in the second year, and the corresponding monthly data is labeled as sample corrosion parameters, which can be refined to the estimated monthly corrosion increment through interpolation method.

[0054] For the task type of the boiler aging predictor, a random forest can be selected to build the boiler aging predictor.

[0055] The input features of the boiler aging predictor are the various boiler operation data in the sample usage data sequence, including operation time, load size, furnace temperature, pipeline pressure, etc.

[0056] The output target of the boiler aging predictor is the sample corrosion parameter, which can be specifically the maximum corrosion depth of the boiler.

[0057] In the parameter setting of the boiler aging predictor, the number of decision trees is 100 to balance the prediction accuracy and calculation efficiency. The maximum tree depth is 15 to limit the complexity of a single tree and avoid overfitting. The minimum split sample size is 5, that is, when the number of node samples is less than 5, the splitting is stopped to ensure the stability of the branches. The minimum leaf node sample size is 2 to ensure that the leaf node has enough samples to support the prediction result. The feature sampling ratio is 0.8, that is, 80% of the input features are randomly selected for each tree to enhance the model generalization ability.

[0058] In the training of the boiler aging predictor, not less than 500 groups of sample data are selected from the sample usage data sequence set and the sample corrosion parameter set for training the boiler aging predictor, wherein each group of samples contains 1 usage data sequence and 1 corresponding labeled corrosion parameter. The samples cover different running time lengths, different load fluctuation ranges and different working conditions of the boiler running scene to ensure data diversity. 20% of the sample data is divided from the not less than 500 groups of sample data as a validation set. The number of training rounds is set to 50 rounds, and each round of training is a complete training of all 500 groups of samples, that is, the model parameters are updated by traversing all the samples.

[0059] When the prediction accuracy of the model on the validation set fluctuates by no more than 1% in the last 5 rounds of training, and the average value of the absolute value of the difference between the actual corrosion parameter and the predicted corrosion parameter is stable within 0.02 mm, it is determined that the model converges to obtain the boiler aging predictor.

[0060] Finally, the usage data sequence is input into the boiler aging predictor, and the predicted corrosion parameter is output by inputting the usage data sequence into the boiler aging predictor. For example, the running time, load size, furnace temperature, pipe pressure and other usage data sequences of a certain boiler are input into the boiler aging predictor, and the predicted corrosion depth is output as 0.12 mm.

[0061] In step S300 of the embodiments of the present application, the conductivity correction coefficient and the phosphate correction coefficient are generated according to the predicted corrosion parameter, the conductivity and the phosphate parameter are corrected to obtain the corrected conductivity and the corrected phosphate parameter, including:

[0062] The predicted corrosion parameter is input into the parameter correction classification table to output the conductivity correction coefficient and the phosphate correction coefficient, wherein the parameter correction classification table includes a conductivity correction classification sub-table and a phosphate correction classification sub-table, and is constructed based on the mapping relationship between the sample corrosion parameter set and the sample conductivity error coefficient set and the sample phosphate error coefficient set in the boiler running history data.

[0063] The predicted corrosion parameter is input into the parameter correction classification table to map and classify the conductivity correction coefficient and the phosphate correction coefficient;

[0064] The conductivity correction coefficient and the phosphate correction coefficient are used to correct and calculate the conductivity and the phosphate parameter respectively to obtain the corrected conductivity and the corrected phosphate parameter.

[0065] In the actual boiler running process, after the boiler is corroded, iron ions and colored complexes are generated, which affect the accuracy of the conductivity measurement and the accuracy of the spectrophotometric test of the phosphate

[0066] Therefore, the purpose of step S300 of the embodiment of the present application is to eliminate the conductivity and phosphate parameter measurement errors caused by boiler corrosion by generating a phosphate correction coefficient based on the predicted corrosion parameter, so that the two key boiler water parameters are more accurate, reliable data basis is provided for subsequent phosphate addition optimization, and the pertinence and effectiveness of boiler water treatment and corrosion protection are ensured.

[0067] Firstly, a parameter correction classification table is needed to be constructed, which includes a conductivity correction classification sub-table and a phosphate correction classification sub-table. The construction basis is the mapping relationship between the sample corrosion parameter set and the sample conductivity error coefficient set and the sample phosphate error coefficient set in the boiler operation history data.

[0068] Specifically, the error coefficient under different sample corrosion parameters is needed to be analyzed by analyzing the deviation of the actual conductivity and phosphate parameters from the ideal value under different corrosion degrees in the boiler operation history data, that is, the error coefficient under different sample corrosion parameters, and the mapping relationship between the sample corrosion parameter set and the sample conductivity error coefficient set and the sample phosphate error coefficient set in the boiler operation history data is constructed.

[0069] For example, under a certain sample corrosion parameter condition, the actual measured value of the conductivity is 2% lower than the true value without corrosion interference, that is, the error coefficient is -2%, and the corresponding conductivity correction coefficient is 1.02; the actual measured value of the phosphate is 1% higher than the true value without corrosion interference, that is, the error coefficient is 1%, and the corresponding phosphate correction coefficient is 0.99.

[0070] Further, the corresponding relationship between the corrosion parameter and the correction coefficient is established to form the parameter correction classification table, for example, the corrosion depth of 0.1 mm corresponds to the conductivity correction coefficient of 1.02 and the phosphate correction coefficient of 0.98.

[0071] Then, the predicted corrosion parameter is input into the parameter correction classification table, and the conductivity correction coefficient and the phosphate correction coefficient are obtained by mapping classification. For example, the predicted corrosion parameter is the corrosion depth of 0.15 mm, the parameter correction classification table is consulted, it is found that the corrosion depth corresponds to the conductivity correction coefficient of 1.05 and the phosphate correction coefficient of 0.95, if the original conductivity collected in step S100 is 300 μS / cm and the original phosphate parameter is 5 mg / L, then the corrected conductivity = 300 x 1.05 = 315 μS / cm; the corrected phosphate parameter = 5 x 0.95 = 4.75 mg / L. The corrected parameters here are closer to the real state of the boiler water and can be used for subsequent phosphate addition optimization.

[0072] In step S400 of the embodiment of the present application, the latest use data is extracted from the use data sequence, the boiler water characteristics are predicted, the predicted conductivity and the predicted phosphate parameter are obtained, the similarity with the corrected conductivity and the corrected phosphate parameter is analyzed, and the parameter weight and the corrosion weight are configured, including:

[0073] extracting latest usage data in the usage data sequence;

[0074] inputting the latest usage data into a boiler water feature predictor to output a predicted conductivity and a predicted phosphate parameter, wherein the boiler water feature predictor is trained by a sample usage data set, a sample predicted conductivity set and a sample predicted phosphate parameter set;

[0075] calculating an average similarity of the corrected conductivity and the corrected phosphate parameter with the predicted conductivity and the predicted phosphate parameter;

[0076] taking the average similarity as a parameter weight and calculating a corrosion weight.

[0077] In the embodiments of the present application, the above steps are aimed at predicting the future features of the boiler water by extracting the latest usage data, comparing the similarity of the latest usage data with the corrected conductivity and the corrected phosphate parameter to determine the parameter weight, and determining the corrosion weight in combination with the corrosion condition, so as to provide a quantitative weight basis for subsequent phosphate addition optimization, make the optimization results not only fit the actual state of the boiler water, but also fully consider the corrosion protection demand, and improve the accuracy and pertinence of phosphate addition.

[0078] Firstly, a group of usage data closest to the current time, i.e. the latest usage data, such as the running data of the last 1 hour, containing running time, load size, furnace temperature, pipe pressure, etc., is selected from the boiler usage data sequence collected in step S100 as an input basis reflecting the latest running state of the boiler.

[0079] Considering the task type of the boiler water feature predictor, a gradient boosting regression tree can be used to build it.

[0080] The input features of the boiler water feature predictor are the latest usage data, and the output targets are the sample predicted conductivity and the sample predicted phosphate parameter. In the parameter setting of the boiler water feature predictor, the number of basic decision trees is 200. The learning rate is 0.05, which is used to control the contribution degree of each tree to the model to avoid overfitting. The maximum tree depth is 8, which is used to limit the complexity of a single tree to prevent the model from overfitting the training data. The minimum number of splitting samples is 10, i.e. the node sample number is less than 10, which stops splitting to ensure the stability of the branch. The subsampling ratio is 0.9, i.e. 90% of the samples are randomly selected for training for each tree to enhance the generalization ability of the model.

[0081] The training data of the boiler water feature predictor is a dataset, a sample predicted conductivity set and a sample predicted phosphate parameter set, and the amount of data is not less than 800 groups, and 20% of the training data is divided as a validation set. The number of training rounds is set to 100 rounds, and each round of training is a complete training of all 800 samples, that is, the model parameters are updated by traversing all samples. When the prediction error of the boiler water feature predictor on the validation set meets the conductivity prediction error ≤3 μS / cm, the phosphate parameter prediction error ≤0.1 mg / L, and the error fluctuation amplitude is not more than 5% in 8 continuous training, it is determined that the boiler water feature predictor converges. The latest use data is input into the boiler water feature predictor, and the predicted conductivity and predicted phosphate parameter are obtained.

[0082] For example, if the input past 1 hour running time is 60 minutes, the average load is 90%, the furnace temperature is 185°C, and the pipeline pressure is 1.2 MPa, the predicted conductivity is 320 μS / cm, and the predicted phosphate parameter is 4.8 mg / L.

[0083] Further, the average similarity of the corrected conductivity and the corrected phosphate parameter with the predicted conductivity and the predicted phosphate parameter needs to be calculated. First, the similarity of the two parameters needs to be calculated, for example, the similarity of parameters A and B can be calculated as similarity = [1- | parameter A- parameter B| / max (parameter A, parameter B)] x 100%, wherein parameter A is the corrected value, such as the corrected conductivity, parameter B is the predicted value, such as the predicted conductivity, | parameter A- parameter B| represents the absolute difference between the two, and max (parameter A, parameter B) represents the maximum of the two.

[0084] Suppose the corrected conductivity obtained in step S300 is 315 μS / cm, the corrected phosphate parameter is 4.75 mg / L, the predicted conductivity is 320 μS / cm, and the predicted phosphate parameter is 4.8 mg / L. According to the above formula, the similarity of the corrected conductivity and the predicted conductivity is 98%, the similarity of the corrected phosphate parameter and the predicted phosphate parameter is 97%, and the average similarity of the two groups of parameters is (98%+97%) / 2=97.5%. Then the average similarity is taken as the parameter weight, that is, the parameter weight is 97.5%, that is, 0.975.

[0085] Further, the corrosion weight needs to be calculated according to the preset rule, for example, when the corrosion depth is <0.05 mm, the corresponding corrosion weight is 0.3; when the corrosion depth is >0.3 mm, the corresponding corrosion weight is 1.0, and so on.

[0086] In step S400 of the embodiment of the present application, the predicted corrosion parameter is combined to optimize the phosphate addition, and the optimized phosphate addition parameter is obtained, and the phosphate is added, including:

[0087] The first phosphate addition parameter is randomly generated;

[0088] According to the predicted conductivity, the predicted phosphate parameter, the corrected conductivity and the corrected phosphate parameter, a fused conductivity and a fused phosphate parameter are calculated and obtained;

[0089] According to the pH, the fused conductivity and the fused phosphate parameter, a basic phosphate concentration is processed and obtained, a first added phosphate concentration is calculated in combination with the first phosphate addition parameter, and a similarity with a standard phosphate concentration is calculated to obtain a first addition fitness;

[0090] According to the first added phosphate concentration in combination with the predicted corrosion parameter, a first corrosion protection coefficient is obtained by corrosion protection prediction;

[0091] According to the parameter weight and the corrosion weight, the first addition fitness and the first corrosion protection coefficient are weighted and calculated to obtain a first phosphate fitness;

[0092] The iterative optimization of the phosphate addition parameter is continued until convergence, and an optimized phosphate addition parameter with the maximum phosphate fitness is obtained, and the phosphate addition is performed.

[0093] In the embodiments of the present application, the purpose of the step of randomly generating the first phosphate addition parameter and calculating the first addition fitness is to provide an initial evaluation basis for the subsequent optimization of the phosphate addition parameter by randomly generating the first phosphate addition parameter, calculating the basic phosphate concentration in combination with the fused boiler water parameter, and determining the first addition fitness by comparing the similarity between the added phosphate concentration, i.e., the first added phosphate concentration, and the standard phosphate concentration, so as to ensure that the finally obtained optimized parameter can meet the requirements of the boiler water standard and adapt to the corrosion protection demand corresponding to the predicted corrosion parameter.

[0094] Firstly, the first phosphate addition parameter needs to be randomly generated, for example, the phosphate addition amount 0.5 g / L,

[0095] Then, the fused conductivity and the fused phosphate parameter are calculated and obtained according to the predicted conductivity, the predicted phosphate parameter, the corrected conductivity and the corrected phosphate parameter.

[0096] The fused conductivity is weighted and fused based on the parameter weight and the corrosion weight, i.e., fused conductivity=(parameter weight×corrected conductivity+corrosion weight×predicted conductivity) / (parameter weight+corrosion weight). For example, the parameter weight is 0.975, the corrosion weight is 0.8, the predicted conductivity is 320 μS / cm, and the corrected conductivity is 315 μS / cm, then fused conductivity=(0.975×315+0.8×320) / (0.975+0.8)≈317 μS / cm.

[0097] Similarly, the fused phosphate parameter is obtained by weighting the predicted phosphate parameter and the corrected phosphate parameter based on the parameter weight and the corrosion weight, such as fused phosphate parameter=(parameter weight×corrected phosphate parameter+corrosion weight×predicted phosphate parameter) / (parameter weight+corrosion weight). For example, the predicted phosphate parameter is 4.8 mg / L, and the corrected phosphate parameter is 4.75 mg / L, then the fused phosphate parameter=(0.975×4.75+0.8×4.8) / (0.975+0.8)≈4.77 mg / L.

[0098] Further, the base phosphate concentration is obtained by processing the pH, the fused conductivity, and the fused phosphate parameter, the first added phosphate concentration is obtained by combining the first phosphate addition parameter, and the similarity to the standard phosphate concentration is calculated to obtain the first addition fitness.

[0099] The base phosphate concentration is obtained by combining the pH parameter collected in step S100, the fused conductivity, and the fused phosphate parameter, and a boiler water chemical property model such as a phosphate buffer solution equilibrium model, and reflects the base sulfate concentration state of the current boiler water without adding phosphate. The application mode of the boiler water chemical property model is a prior art, which is not described here. For example, according to the pH=9.2, the fused conductivity 317 μS / cm, and the fused phosphate parameter 4.77 mg / L, the base phosphate concentration is calculated to be 4.77 mg / L.

[0100] The first added phosphate concentration is obtained by adding the base phosphate concentration and the first phosphate addition parameter to obtain the predicted phosphate concentration after addition, such as in the above example, the first added phosphate concentration=4.77+0.5=5.27 mg / L.

[0101] The first addition fitness is the similarity between the first added phosphate concentration and the preset standard phosphate concentration, wherein the preset standard phosphate concentration is an ideal concentration that meets the safe operation of the boiler, such as 5.0 mg / L. The calculation method of the similarity between the two can refer to the calculation method of the similarity between the parameters A and B described above. According to the method, the similarity between the first added phosphate concentration and the preset standard phosphate concentration is 95%, and the first addition fitness is 95%, i.e. 0.95. The higher the first addition fitness, the closer the first addition fitness corresponding to the phosphate addition parameter to the ideal state.

[0102] In step S400 of the embodiments of the present application, the first corrosion protection coefficient is obtained by corrosion protection prediction based on the first added phosphate concentration and the predicted corrosion parameter, including:

[0103] According to the boiler phosphoric acid addition data in the historical time, a sample corrosion parameter set and a sample phosphoric acid radical concentration set are collected, and the scale ratio of the protective layer in the boiler under different sample corrosion parameters and sample phosphoric acid radical concentrations is tested to obtain a sample corrosion protection coefficient set;

[0104] A machine learning is used to construct a corrosion protection predictor, which is supervised trained based on the sample corrosion parameter set, the sample phosphoric acid radical concentration set and the sample corrosion protection coefficient set until the requirement is met.

[0105] The first added phosphoric acid radical concentration and the predicted corrosion parameter are input into the corrosion protection predictor to obtain a first corrosion protection coefficient.

[0106] In the embodiments of the present application, the above steps are used to construct and use the corrosion protection predictor, and the corrosion protection effect of the boiler under the first added phosphoric acid radical concentration is quantitatively evaluated by combining the fitness of the first added phosphoric acid radical concentration and the predicted corrosion parameter, i.e., the first corrosion protection coefficient, which provides an evaluation basis for the corrosion protection layer for the optimization of the subsequent phosphoric acid salt addition parameters, and ensures that the finally optimized phosphoric acid salt addition parameters can meet the boiler water concentration standard and effectively improve the corrosion protection capability.

[0107] In this step, the protective layer in the boiler refers to a thin film or covering layer formed on the surface of the inner wall of the boiler metal, which is mainly generated by the chemical reaction between the phosphates in the boiler water and the metal surface, can isolate the inner wall of the boiler metal from the boiler water, prevent the corrosive substances in the boiler water from contacting the metal surface, and reduce the occurrence of electrochemical corrosion and chemical corrosion from the root, thereby reducing the risk of thinning and perforation of the boiler pipe wall.

[0108] In this embodiment, first, according to the boiler phosphoric acid addition data in the historical time, a sample corrosion parameter set and a sample phosphoric acid radical concentration set are collected. Then, the scale ratio of the protective layer in the boiler under different sample corrosion parameters and sample phosphoric acid radical concentrations is tested to obtain a sample corrosion protection coefficient set

[0109] For example, in the historical sample, when the sample corrosion parameter is a corrosion depth of 0.1 mm and the sample phosphate concentration is 5.0 mg / L, it is detected that the coverage ratio of the protective layer in the boiler is 85%, and the labeled sample corrosion protection coefficient is 0.85; when the sample corrosion depth is 0.1 mm and the sample phosphate concentration is 5.5 mg / L, the coverage ratio of the protective layer in the boiler is 90%, and the labeled coefficient is 0.9, and so on. The above-mentioned method is used to obtain the sample corrosion parameter set, the sample phosphate concentration set and the sample corrosion protection coefficient set, which are used as the training sample data set of the corrosion protection predictor, and 20% of the data in the training sample data set is used as the verification set. The training sample data set needs to include not less than 600 groups of training sample data, and each group of training sample data needs to include 1 sample corrosion parameter, 1 sample phosphate concentration and 1 sample corrosion protection coefficient.

[0110] Further, a machine learning is needed to build a corrosion protection predictor, which is supervised trained based on the sample corrosion parameter set, the sample phosphate concentration set and the sample corrosion protection coefficient set to meet the requirements

[0111] According to the task type of the corrosion protection predictor, a multi-layer perception can be selected to build the corrosion protection predictor.

[0112] Specifically, the input features of the corrosion protection predictor are corrosion parameters and phosphate concentrations; and the output target is a corrosion protection coefficient.

[0113] In the parameter setting of the corrosion protection predictor, it is a 3-layer structure, including 1 input layer, 1 hidden layer and 1 output layer. The number of hidden layer neurons is 32. The hidden layer uses the ReLU function, and the output layer uses the Sigmoid function. The learning rate is 0.01 to control the parameter update speed and avoid convergence shock. The batch size is 32.

[0114] In the training of the corrosion protection predictor, the number of training rounds is set to 80 rounds, and each round is a complete training of all 600 samples, that is, all samples are traversed to update the network parameters. When the prediction error of the model on the verification set, that is, the average value of the absolute difference between the predicted protection coefficient and the actual labeled coefficient, is ≤0.02 in the last 6 consecutive training rounds, and the error fluctuation amplitude is not more than 3%, the model is determined to be converged, and the corrosion protection predictor is obtained.

[0115] The first added phosphate concentration and the predicted corrosion parameter are input into the corrosion protection predictor, and the first corrosion protection coefficient is obtained.

[0116] For example, if the first added phosphate concentration is 5.27 mg / L and the predicted corrosion parameter is a corrosion depth of 0.12 mm, after inputting both into the corrosion protection predictor, the corrosion protection predictor outputs a first corrosion protection coefficient of 0.88, indicating that the boiler corrosion protection effect is good under this phosphate addition parameter, but there is still room for optimization.

[0117] Furthermore, the first addition fitness and the first corrosion protection coefficient need to be weighted and calculated based on the parameter weight and corrosion weight to obtain the first phosphoric acid fitness. Specifically, the first addition fitness and the first corrosion protection coefficient need to be weighted based on the parameter weight and corrosion weight. The calculation formula is: First phosphoric acid fitness = (Parameter weight × First addition fitness) + (Corrosion weight × First corrosion protection coefficient).

[0118] For example, if the parameter weight is 0.975, the corrosion weight is 0.8, the first addition fitness is 0.95, and the first corrosion protection coefficient is 0.88, then the first phosphoric acid fitness = (0.975 × 0.95) + (0.8 × 0.88) ≈ 0.926 + 0.704 = 1.63.

[0119] Furthermore, iterative optimization of the phosphate addition parameters needs to be continued until convergence, obtaining the optimized phosphate addition parameters with the highest phosphate fitness, and then phosphate addition can be performed.

[0120] Optionally, the iterative optimization process can be based on the first phosphate addition parameter, by increasing or decreasing the concentration of added phosphate, to generate new phosphate addition parameters, namely the second, third... nth phosphate addition parameters.

[0121] For each new phosphate addition parameter, its corresponding addition fitness, corrosion protection coefficient, and phosphate fitness are repeatedly calculated in the same way as the first phosphate addition parameter.

[0122] Then, the phosphate addition parameters are continuously adjusted iteratively until the difference in phosphate fitness between two adjacent iterations is less than a preset threshold. If the difference is ≤0.5%, convergence is determined.

[0123] The phosphate addition parameter with the highest phosphate fitness during the iteration process is selected as the optimized phosphate addition parameter, and phosphate addition is performed. For example, the optimized phosphate addition parameter can be 0.55 g / L.

[0124] Example 2, as Figure 2 As shown, based on the same inventive concept as the intelligent phosphate addition method for low-phosphorus boiler water provided in Embodiment 1, this embodiment of the invention also provides an intelligent phosphate addition system for low-phosphorus boiler water, comprising:

[0125] A key data acquisition module 11 is configured to test and acquire conductivity, pH, and phosphate parameters in the boiler water, and to acquire a usage data sequence of the boiler;

[0126] A predicted corrosion parameter acquisition module 12 is configured to perform boiler corrosion prediction based on the usage data sequence to obtain a predicted corrosion parameter;

[0127] A measured parameter correction module 13 is configured to generate a conductivity correction coefficient and a phosphate correction coefficient based on the predicted corrosion parameter, correct the conductivity and phosphate parameters, and obtain corrected conductivity and corrected phosphate parameters;

[0128] A phosphate addition optimization module 14 is configured to extract the latest usage data from the usage data sequence, perform boiler water feature prediction to obtain predicted conductivity and predicted phosphate parameters, analyze the similarity with the corrected conductivity and corrected phosphate parameters, configure a parameter weight and a corrosion weight, combine the predicted corrosion parameter to perform phosphate addition optimization, obtain optimized phosphate addition parameters, and perform phosphate addition.

[0129] Further, the key data acquisition module 11 includes the following execution steps:

[0130] Test and acquire conductivity, pH, and phosphate parameters in the boiler water;

[0131] Acquire a usage data sequence of the boiler after the boiler is put into use, wherein each usage data includes boiler operation data in a usage time period.

[0132] Further, the predicted corrosion parameter acquisition module 12 includes the following execution steps:

[0133] Acquire a boiler aging predictor;

[0134] Input the usage data sequence into the boiler aging predictor, input into the boiler aging predictor, and output to obtain a predicted corrosion parameter.

[0135] The acquisition step of the boiler aging predictor includes:

[0136] According to the boiler operation data in a historical time, acquire a sample usage data sequence set, and label the corrosion size of the boiler under different sample usage data sequences to obtain a sample corrosion parameter set;

[0137] Take the usage data sequence as input data and the corrosion parameter as output data to construct a boiler aging predictor based on machine learning;

[0138] Use the sample usage data sequence set and the sample corrosion parameter set to supervise the training of the boiler aging predictor until the accuracy converges.

[0139] Further, the measurement parameter correction module 13 comprises the following execution steps:

[0140] inputting the predicted corrosion parameter into a parameter correction classification table, and outputting obtained conductivity correction coefficients and phosphate correction coefficients, wherein the parameter correction classification table comprises a conductivity correction classification sub-table and a phosphate correction classification sub-table, and is constructed based on a mapping relationship between a sample corrosion parameter set and a sample conductivity error coefficient set and a sample phosphate error coefficient set in the boiler operation history data;

[0141] inputting the predicted corrosion parameter into the parameter correction classification table, and mapping and classifying to obtain conductivity correction coefficients and phosphate correction coefficients;

[0142] correcting and calculating the conductivity and the phosphate parameter respectively by using the conductivity correction coefficients and the phosphate correction coefficients, to obtain corrected conductivity and corrected phosphate parameters.

[0143] Further, the phosphate addition optimization module 14 comprises the following execution steps:

[0144] extracting the latest use data in the use data sequence;

[0145] inputting the latest use data into a boiler water feature predictor, and outputting obtained predicted conductivity and predicted phosphate parameters, wherein the boiler water feature predictor is trained by using a sample use data set, a sample predicted conductivity set, and a sample predicted phosphate parameter set;

[0146] calculating the average similarity of the corrected conductivity and the corrected phosphate parameters and the predicted conductivity and the predicted phosphate parameters;

[0147] taking the average similarity as a parameter weight, and calculating to obtain a corrosion weight.

[0148] randomly generating a first phosphate addition parameter;

[0149] calculating to obtain fused conductivity and fused phosphate parameters according to the predicted conductivity, the predicted phosphate parameter, the corrected conductivity, and the corrected phosphate parameter;

[0150] processing to obtain a basic phosphate concentration according to the pH, the fused conductivity, and the fused phosphate parameter, calculating to obtain a first added phosphate concentration in combination with the first phosphate addition parameter, and calculating the similarity with a standard phosphate concentration to obtain a first added fitness;

[0151] performing corrosion protection prediction according to the first added phosphate concentration in combination with the predicted corrosion parameter, to obtain a first corrosion protection coefficient;

[0152] According to the parameter weight and the corrosion weight, the first adding fitness and the first corrosion protection coefficient are weighted and calculated to obtain a first phosphoric acid fitness;

[0153] The iterative optimization of the phosphating parameter is continuously performed until convergence, an optimized phosphating parameter with the maximum phosphoric acid fitness is obtained, and phosphating is performed.

[0154] The first corrosion protection coefficient is obtained by performing corrosion protection prediction according to the first adding phosphate concentration and the predicted corrosion parameter, and the corrosion protection prediction includes:

[0155] According to the phosphoric acid adding data of the boiler in a historical time, a sample corrosion parameter set and a sample phosphate concentration set are collected, and the size ratio of the protective layer in the boiler under different sample corrosion parameters and sample phosphate concentrations is tested and obtained, and a sample corrosion protection coefficient set is labeled;

[0156] A machine learning is used to construct a corrosion protection predictor, and supervised training is performed based on the sample corrosion parameter set, the sample phosphate concentration set and the sample corrosion protection coefficient set until a requirement is met;

[0157] The first adding phosphate concentration and the predicted corrosion parameter are input into the corrosion protection predictor, and the first corrosion protection coefficient is output.

[0158] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0159] Those skilled in the art should understand that the embodiments of the present application can provide a method, a system or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0160] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams can be implemented by computer program instructions, and the combination of flows and / or blocks in the flowcharts and / or block diagrams. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a machine that implements the flowcharts and / or block diagrams. Figure 1 one flow or multiple flows and / or blocksFigure 1 means for performing the function specified in the block or blocks.

[0161] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions Figure 1 one or more flowcharts and / or blocks Figure 1 means for performing the function specified in the block or blocks.

[0162] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions Figure 1 one or more flowcharts and / or blocks Figure 1 means for performing the function specified in the block or blocks.

[0163] Although preferred embodiments of the application have been described, those skilled in the art will recognize that additional modifications and variations may be possible in light of the above teachings.

[0164] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application and its equivalents.

Claims

1. A phosphate intelligent addition process for low phosphorus boiler water, characterized by, The method comprises: Test and collect conductivity, pH, and phosphate parameters in the boiler water, and collect a usage data sequence of the boiler; According to the usage data sequence, perform boiler corrosion prediction to obtain a predicted corrosion parameter; According to the predicted corrosion parameter, generate a conductivity correction coefficient and a phosphate correction coefficient, correct the conductivity and the phosphate parameter, and obtain a corrected conductivity and a corrected phosphate parameter; Extract the latest usage data from the usage data sequence, perform boiler water feature prediction to obtain a predicted conductivity and a predicted phosphate parameter, analyze the similarity with the corrected conductivity and the corrected phosphate parameter, configure a parameter weight and a corrosion weight, combine the predicted corrosion parameter to perform phosphate addition optimization, obtain an optimized phosphate addition parameter, and perform phosphate addition, including: Extract the latest usage data from the usage data sequence; Input the latest usage data into a boiler water feature predictor to output a predicted conductivity and a predicted phosphate parameter, wherein the boiler water feature predictor is trained using a sample usage data set, a sample predicted conductivity set, and a sample predicted phosphate parameter set; Calculate the average similarity of the corrected conductivity and the corrected phosphate parameter with the predicted conductivity and the predicted phosphate parameter; Take the average similarity as the parameter weight, and calculate a corrosion weight; Randomly generate a first phosphate addition parameter; According to the predicted conductivity, the predicted phosphate parameter, the corrected conductivity, and the corrected phosphate parameter, calculate a fused conductivity and a fused phosphate parameter; According to the pH, the fused conductivity, and the fused phosphate parameter, process a basic phosphate concentration, combine the first phosphate addition parameter to calculate a first added phosphate concentration, and calculate the similarity with a standard phosphate concentration to obtain a first addition fitness; According to the first added phosphate concentration, perform corrosion protection prediction combined with the predicted corrosion parameter to obtain a first corrosion protection coefficient; According to the parameter weight and the corrosion weight, perform weighted calculation on the first addition fitness and the first corrosion protection coefficient to obtain a first phosphate fitness. Continue to perform iterative optimization of the phosphate addition parameter until convergence, obtain an optimized phosphate addition parameter with the maximum phosphate fitness, and perform phosphate addition.

2. The phosphate smart addition process for low phosphorus boiler water as claimed in claim 1 wherein, Test and collect conductivity, pH, and phosphate parameters in the boiler water, and collect a usage data sequence of the boiler, including: Test and collect conductivity, pH, and phosphate parameters in the boiler water; Obtain a usage data sequence after the boiler is put into use, wherein each usage data comprises boiler operation data in a usage time period.

3. The phosphate smart addition process for low phosphorus boiler water as claimed in claim 1 wherein, According to the usage data sequence, perform boiler corrosion prediction to obtain a predicted corrosion parameter, including: Obtain a boiler aging predictor; Input the usage data sequence into the boiler aging predictor, input into the boiler aging predictor, and output a predicted corrosion parameter.

4. The phosphate smart addition process for low phosphorus boiler water as claimed in claim 3 wherein, The obtaining step of the boiler aging predictor comprises: According to the boiler operation data in the historical time, collect a sample usage data sequence set, and label the corrosion size of the boiler under different sample usage data sequences to obtain a sample corrosion parameter set; The boiler aging predictor based on machine learning is constructed by taking the use data sequence as input data and the corrosion parameter as output data; The boiler aging predictor is supervised trained by using the sample use data sequence set and the sample corrosion parameter set until the accuracy converges.

5. The phosphate smart addition process for low phosphorus boiler water as claimed in claim 1 wherein, The conductivity correction coefficient and the phosphate correction coefficient are generated according to the predicted corrosion parameter, the conductivity and the phosphate parameter are corrected, and the corrected conductivity and the corrected phosphate parameter are obtained, including: The predicted corrosion parameter is input into the parameter correction classification table, and the conductivity correction coefficient and the phosphate correction coefficient are obtained by mapping and classifying, wherein the parameter correction classification table includes a conductivity correction classification sub-table and a phosphate correction classification sub-table, and is constructed based on the mapping relationship between the sample corrosion parameter set and the sample conductivity error coefficient set and the sample phosphate error coefficient set in the boiler operation history data; The predicted corrosion parameter is input into the parameter correction classification table, and the conductivity correction coefficient and the phosphate correction coefficient are obtained by mapping and classifying, wherein the parameter correction classification table includes a conductivity correction classification sub-table and a phosphate correction classification sub-table, and is constructed based on the mapping relationship between the sample corrosion parameter set and the sample conductivity error coefficient set and the sample phosphate error coefficient set in the boiler operation history data; The conductivity correction coefficient and the phosphate correction coefficient are respectively used for correcting and calculating the conductivity and the phosphate parameter, and the corrected conductivity and the corrected phosphate parameter are obtained.

6. The phosphate smart-dosing method for low phosphorus boiler water as claimed in claim 1, wherein, The first corrosion protection coefficient is obtained by combining the first added phosphate concentration with the predicted corrosion parameter for corrosion protection prediction, including: According to the boiler phosphate adding data in the historical time, the sample corrosion parameter set and the sample phosphate concentration set are collected, and the scale ratio of the protective layer in the boiler under different sample corrosion parameters and sample phosphate concentrations is tested and obtained, and the sample corrosion protection coefficient set is labeled; The corrosion protection predictor is constructed by using machine learning, and is supervised trained based on the sample corrosion parameter set, the sample phosphate concentration set and the sample corrosion protection coefficient set until the requirements are met; The first added phosphate concentration and the predicted corrosion parameter are input into the corrosion protection predictor, and the first corrosion protection coefficient is obtained by output.

7. A phosphate intelligent addition system for low phosphorus boiler water, characterized by, The system includes: A key data acquisition module is used for testing and collecting the conductivity, pH and phosphate parameters in the boiler water, and collecting the use data sequence of the boiler; A predicted corrosion parameter acquisition module is used for predicting the corrosion of the boiler according to the use data sequence, and obtaining the predicted corrosion parameter; A measurement parameter correction module is used for generating the conductivity correction coefficient and the phosphate correction coefficient according to the predicted corrosion parameter, correcting the conductivity and the phosphate parameter, and obtaining the corrected conductivity and the corrected phosphate parameter; A phosphate addition optimization module is used for extracting the latest use data in the use data sequence, predicting the characteristics of the boiler water, obtaining the predicted conductivity and the predicted phosphate parameter, analyzing the similarity with the corrected conductivity and the corrected phosphate parameter, configuring the parameter weight and the corrosion weight, combining the predicted corrosion parameter for phosphate addition optimization, obtaining the optimized phosphate addition parameter, and adding phosphate, including: The latest use data is extracted in the use data sequence; inputting the latest use data into a boiler water characteristic predictor to output predicted conductivity and predicted phosphate parameters, wherein the boiler water characteristic predictor is trained by using a sample use data set, a sample predicted conductivity set and a sample predicted phosphate parameter set; calculating an average similarity of the corrected conductivity and corrected phosphate parameters and the predicted conductivity and predicted phosphate parameters; using the average similarity as a parameter weight and calculating a corrosion weight; randomly generating a first phosphate addition parameter; calculating a fused conductivity and a fused phosphate parameter according to the predicted conductivity, the predicted phosphate parameter, the corrected conductivity and the corrected phosphate parameter; calculating a base phosphate concentration according to the pH, the fused conductivity and the fused phosphate parameter, calculating a first added phosphate concentration by combining the first phosphate addition parameter, and calculating a similarity with a standard phosphate concentration to obtain a first addition fitness; conducting corrosion protection prediction according to the first added phosphate concentration combined with the predicted corrosion parameter to obtain a first corrosion protection coefficient; weighting and calculating the first addition fitness and the first corrosion protection coefficient according to the parameter weight and the corrosion weight to obtain a first phosphate fitness; continuing iteration optimization of the phosphate addition parameter until convergence to obtain an optimized phosphate addition parameter with the maximum phosphate fitness, and adding phosphate.

Images