High-precision wide-range slurry chloride ion online monitoring system and method based on dynamic self-adaption

Abstract

The invention belongs to the technical field of wet desulphurization, and particularly relates to a high-precision wide-range slurry chloride ion online monitoring system and method based on dynamic self-adaption, and the system comprises a reactor, a slurry storage tank, a water storage tank, a chloride ion correction unit and an upper computer. Slurry and dilution water are quantitatively injected into the reactor through a communication assembly controlled by a peristaltic pump. A chloride ion electrode, a temperature sensor, a pH detector, a double-layer stirrer and a sprayer are integrated in the reactor. The correction unit injects standard liquid through a peristaltic pump. And the upper computer is connected with and cooperatively controls all the components and executes a self-adaptive detection process. The concentration value of background chloride ions is measured, then the mixed value is measured, and according to comparison of threshold values, the concentration is calculated, or the dilution specific gravity is adjusted for measurement, or direct measurement is carried out; and finally, inverting the actual concentration. The system realizes on-line measurement, the dilution ratio is judged and adjusted, the operation stability is ensured through the detection and correction functions, and the monitoring efficiency and reliability are improved.

Description

Technical Field

[0001] This invention belongs to the field of wet desulfurization technology, specifically relating to a high-precision, wide-range online monitoring system and method for chloride ions in slurry based on dynamic adaptive technology. Background Technology

[0002] In the wet desulfurization process of coal-fired power plants, the chloride ion concentration in the desulfurization slurry is a key parameter affecting system corrosion, limestone dissolution, wastewater discharge, desulfurization efficiency, and equipment lifespan.

[0003] When performing online monitoring of chloride ions in slurry, related technologies encounter fluctuations in the chloride ion concentration of industrial slurry. When the slurry concentration is low, the concentration of the diluted mixture is easily lower than the effective measurement limit of the chloride ion electrode, resulting in unstable measurement signals and large errors. When the slurry concentration is high, the dilution ratio cannot reduce the concentration to the measurement range suitable for the electrode, resulting in measurement saturation and making it difficult to achieve full-range coverage detection of slurry with different concentrations.

[0004] In related technologies, dilution water may contain trace amounts of chloride ions due to fluctuations in water source or pipeline residues. These background chloride ions will be directly included in the measured value of the mixed solution, leading to an overestimation of the subsequent concentration calculation results, which cannot accurately reflect the chloride ion concentration level of the slurry itself. After the test is completed, the reactor is not rinsed, and residues of the previous batch of slurry or mixed solution can easily remain on the reactor inner wall, electrode probe, and agitator. During subsequent tests, these residues mix with new materials, causing contamination of the slurry or mixed solution, which deviates from the true value and affects the accuracy of the test. Summary of the Invention

[0005] This invention provides a high-precision, wide-range online monitoring system for chloride ions in slurry based on dynamic adaptation. The system can adapt to complex slurry environments and has wide range, high precision, adaptive adjustment, and online automatic calibration functions. It improves the intelligence level of wet desulfurization processes, ensures long-term safe and stable operation of the system, realizes online monitoring of chloride ions and precise quantitative discharge of desulfurization wastewater, and ensures stable and efficient operation of the desulfurization system.

[0006] The system includes: a reactor, a slurry storage tank, a water storage tank, and a host computer; The slurry storage tank is connected to the reactor via the first communication control component; The water storage tank is connected to the reactor via a second communication control component; The reactor is equipped with a chloride ion electrode, a temperature sensor, a double-layer stirrer, a pH meter, and a sprayer. The host computer is electrically connected to the first communication control component, the second communication control component, the pH meter, the chloride ion electrode, the temperature sensor, the double-layer stirrer, and the sprayer, respectively. The host computer is used to obtain the actual chloride ion concentration of the slurry.

[0007] It should be further noted that it also includes: a chloride ion correction unit; The chloride ion correction unit is connected to the reactor; The chloride ion correction unit is equipped with a standard chloride ion solution storage tank, which is connected to the reactor via a third peristaltic pump; The first communication control component is provided with a first flexible hose, and a first peristaltic pump is installed on the first flexible hose; The second communication control component is provided with a second hose, and a second peristaltic pump is installed on the second hose.

[0008] It should be further noted that a level switch is installed on the side wall of the water storage tank; A drain pipe is installed at the bottom of the reactor, and a solenoid valve is installed on the drain pipe; A fixed bracket is installed near the top of the reactor, and the chloride ion electrode, temperature sensor, double-layer stirrer, pH meter and sprayer are fixed to the fixed bracket.

[0009] It should be further explained that the host computer, in conjunction with a long short-term memory network, establishes a multi-parameter correlation prediction model. The model is trained using historical detection data including chloride ion concentration, pH, and temperature. Based on the model, the predicted chloride ion concentration for future time periods is obtained. According to the comparison between the predicted information and the actual working conditions, the chloride ion correction unit, the first interconnection control component, the second interconnection control component, the double-layer stirrer, and the sprayer are controlled to adjust the dilution ratio and stirring time, thereby realizing online monitoring of chloride ions.

[0010] This invention also provides a high-precision, wide-range online monitoring method for chloride ions in slurry, the method comprising: S101: The host computer monitors the water level in the storage tank and controls water replenishment through the level switch to keep the water level in the storage tank at the set height. S102: Control the second peristaltic pump to pump a metered amount of dilution water into the reactor, control the chloride ion electrode to measure the background chloride ion concentration value C0, and control the solenoid valve to empty the reactor; S103: Control the first peristaltic pump to pump a quantitative amount of slurry and control the second peristaltic pump to pump a quantitative amount of dilution water into the reactor. After starting the double-layer stirrer to mix, control the chloride ion electrode to measure the concentration value C1 of the diluted mixed chloride ion. S104: Determine whether C1 exceeds the set limit value Cs1: If C1 exceeds the limit value CS1, then C1 is determined to be valid, and step S108 is executed; If C1 does not exceed the limit value CS1, then C1 is determined to be invalid, and step S105 is executed; S105: After controlling the solenoid valve to drain and controlling the sprayer to rinse the reactor, adjust the volume ratio of dilution water to slurry, re-pump the mixture, and measure the concentration value C2 of the diluted mixed chloride ions. S106: Determine whether C2 exceeds the set limit value CS2: If C2 exceeds the limit value CS2, then C2 is determined to be valid, and C2 is used to replace C1, and step S108 is executed; If C2 does not exceed the limit value CS2, then C2 is determined to be invalid, and step S107 is executed; S107: After controlling the solenoid valve to drain and controlling the sprayer to rinse the reactor, control the first peristaltic pump to pump in a quantitative slurry for direct measurement, and measure the chloride ion concentration value C3. Use C3 as the final measurement value and execute step S108. S108: The host computer calculates the actual chloride ion concentration of the slurry based on the effective measured value Cn, the background chloride ion concentration value C0, and the volume parameters. Cn represents the effective value of C1, C2, or C3.

[0011] It should be further noted that in S108, the method for calculating the actual chloride ion concentration in the slurry is as follows: C (actual chloride ion concentration) = [C1 × V (total) - C0 × V (dilution water)] / V (slurry) Where V(total) = V(dilution water) + V(slurry).

[0012] It should be further noted that after S108, the following also includes: S201: The host computer collects the chloride ion concentration, pH value and temperature value output by the chloride ion electrode, pH meter and temperature sensor during the historical period, and simultaneously records the running time of the first peristaltic pump, the running time of the second peristaltic pump and the running time of the double-layer stirrer at the corresponding time as operation tags. S202: The host computer performs outlier removal and missing value interpolation on the collected historical data, normalizes each physical quantity to the [0,1] interval, and extracts the rate of change, local extrema and mean within the sliding time window as feature vectors. S203: The host computer trains a multi-parameter correlation prediction model based on the processed feature vectors, and takes the multi-parameter features of the current and previous N sampling cycles as inputs to output the recommended dilution water volume, slurry volume and double-layer agitator running time for the next detection cycle. S204: Before executing step S102, the host computer calls the trained multi-parameter correlation prediction model to generate the dilution water injection volume, slurry injection volume and stirring time required for this test based on the latest collected chloride ion concentration trend, pH value and temperature value, and sets the start and stop time of the first peristaltic pump and the second peristaltic pump and the running time of the double-layer agitator accordingly. S205: After completing the calculation of the actual chloride ion concentration, the effective concentration value Cn measured in this study is compared with the concentration value predicted by the model in the past, the prediction error is calculated, and the weight parameters of the multi-parameter correlation prediction model are fine-tuned using the error.

[0013] It should be further noted that the training process of the multi-parameter association prediction model includes the following steps: The host computer acquires data from each sensor and constructs the original training set; Preprocess the original training set; Within each sampling period, calculate the statistical characteristics of the sliding window over a past period, such as the average, maximum, minimum, and rate of change, to form a feature vector set. A multi-parameter association prediction model based on a long short-term memory network is constructed and trained using a feature vector set. The multi-parameter association prediction model consists of an input layer, hidden layers, and an output layer. The input layer receives feature vectors, the hidden layer is responsible for learning the correlation between sequences, and the output layer provides recommended values ​​for dilution water volume, slurry volume, and stirring time. Mean squared error is used as the loss function, and the weights are adjusted through batch stochastic gradient descent until the model converges. After initial training, the model is updated online using newly collected data. After each actual measurement, the effective concentration value Cn is compared with the previously predicted value, and the model weights are adjusted according to the error.

[0014] It should be further noted that the following steps are included after S205: S206: The host computer extracts the predicted chloride ion concentration value Cp from the multi-parameter correlation prediction model before the start of this detection cycle; S207: The host computer calculates the difference between the measured effective concentration value Cn and the predicted concentration value Cp to obtain the absolute error ΔC = |Cn|. Cp|, and determine whether ΔC is greater than the preset error threshold ε; S208: If ΔC>ε, the host computer will combine the complete input feature vector X, the true operation parameters Y, and the target output label corresponding to this detection into a new training sample; S209: The host computer adds new training samples to the online fine-tuning queue and performs a single forward and backward propagation update on the weight matrix of the multi-parameter correlation prediction model, using the mean squared error as the loss function. S210: After the weight update is completed, the host computer stores the updated model parameters and releases the cached predicted value Cp, feature vector X and error ΔC to prepare the running space for the next detection cycle.

[0015] It should be further explained that S104 specifically includes the following steps: The fluctuation range of chloride ion concentration in the historical detection data of the host computer and the influence of historical temperature information fed back by the temperature sensor on the chloride ion concentration measurement are used to set the specific value of the limit value CS1. The host computer obtains the diluted mixed chloride ion concentration value C1 obtained from the chloride ion electrode, and at the same time reads the temperature value T currently measured by the temperature sensor and records the corresponding numerical data. The host computer corrects the acquired C1 value based on the pre-set temperature and concentration correction relationship and the current temperature value T, obtaining the corrected concentration value C1c. C1c is then compared with the set CS1 value, and internal logic operations are used to determine whether C1c is greater than CS1. If the comparison result is that C1c is greater than CS1, the host computer will determine that C1 is valid and trigger the instruction to execute step S108; if the comparison result is that C1c is less than or equal to CS1, C1 will be determined to be invalid and the instruction to execute step S105 will be triggered.

[0016] As can be seen from the above technical solutions, the present invention has the following advantages: The high-precision, wide-range online chloride ion monitoring system for slurry provided by this invention addresses the problem of insufficient full-range coverage. It can obtain C1 at an initial ratio and determine its validity using Cs1. If invalid, the dilution ratio is adjusted to obtain C2, and a second determination is made using CS2; if still invalid, it switches to direct slurry measurement to obtain C3, adapting to different slurry concentrations through a three-level detection logic. A step for measuring the background concentration of dilution water is added to the method. When calculating, the host computer deducts the chloride ion content of the dilution water corresponding to C0, ensuring that the calculation result only reflects the chloride ion concentration of the slurry itself.

[0017] The method employs evacuation and flushing processes. For flushing after invalid detection, the flushing time is extended and the flushing effect is enhanced to ensure that there is no residual material inside the reactor and on the surface of the components. The entire process is automatically controlled by a host computer to control the flow rate and runtime of each peristaltic pump, achieving quantitative material delivery. The host computer retrieves measured values, background chloride ion concentration values, and volume parameters, substitutes them into preset formulas to complete the calculation, and no manual intervention is required.

[0018] The present invention sets two-level validity judgment thresholds, CS1 and CS2, to screen C1 and C2 respectively. Only valid measurements are included in the calculation, and invalid measurements trigger corresponding adjustments to ensure that the output data meets the usage requirements. This enables online monitoring of chloride ions and precise quantitative discharge of desulfurization wastewater, ensuring the stable and efficient operation of the desulfurization system. Attached Figure Description

[0019] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 Schematic diagram of a high-precision, wide-range online chloride ion monitoring system for slurry; Figure 2 Flowchart of a high-precision, wide-range online monitoring method for chloride ions in slurry; Figure 3 This is a schematic diagram of the host computer.

[0021] Explanation of reference numerals in the attached figures: 1. Reactor; 2. Chloride ion electrode; 3. Temperature sensor; 4. Double-layer stirrer; 5. pH meter; 6. Sprayer; 7. First hose; 8. Second hose; 9. First peristaltic pump; 10. Second peristaltic pump; 11. Third peristaltic pump; 12. Slurry storage tank; 13. Water storage tank; 14. Liquid level switch; 15. Standard chloride ion solution storage tank; 16. Host computer; 17. Solenoid valve; 18. Fixing bracket; 101. Processor; 102. Memory; 103. Display module; 104. Communication module. Detailed Implementation

[0022] The high-precision, wide-range online chloride ion monitoring system for slurry according to this application will be described in detail below. Specific details, such as particular system structures and technologies, are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details.

[0023] It should be understood that "one or more" as mentioned in this application refers to one, two, or more, and "multiple" as mentioned in this application refers to two or more. In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0024] To facilitate a clear description of the technical solutions of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" do not necessarily imply that they are different.

[0025] The terms "one embodiment" or "some embodiments" used in this application mean that one or more embodiments of this application include the specific features, structures, or characteristics described in that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this application do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1 The diagram shown is a schematic of a high-precision wide-range online chloride ion monitoring system for slurry in a specific embodiment. The high-precision wide-range online chloride ion monitoring system for slurry based on dynamic adaptation includes: a reactor 1, a slurry storage tank 12, a water storage tank 13, a chloride ion correction unit, and a host computer 16.

[0028] In some embodiments, reactor 1 has a conical bottom design to facilitate complete drainage of waste liquid, and multiple sensor mounting interfaces are evenly distributed on the sidewalls. Slurry storage tank 12 is equipped with a heating device and a temperature monitoring module. Water storage tank 13 is equipped with a level switch 14 and is connected to municipal water supply or process water source via pipeline.

[0029] The chloride ion correction unit includes multiple standard chloride ion solution storage tanks 15, which store low-concentration and high-concentration standard solutions respectively. The host computer 16 adopts a touch screen host with a built-in processor and memory, Ethernet, and monitoring and control software, which can realize functions such as parameter setting, data display, and historical record query.

[0030] Optionally, the slurry storage tank 12 stores the desulfurization slurry to be tested, and the water storage tank 13 provides the dilution water required for testing. The materials are precisely delivered to the reactor 1 via corresponding communication control components. The reactor 1 serves as the testing site, where various integrated testing components and actuators perform mixing and testing operations. The chloride ion calibration unit, according to instructions from the host computer 16, periodically injects standard chloride ion solution into the reactor 1 to calibrate the chloride ion electrode 2. The host computer 16 collects real-time operating and testing data from each component, controls the operation of the entire system, and performs tasks such as concentration calculation and result output.

[0031] In this embodiment, the slurry storage tank 12 is connected to the reactor 1 through a first communication control component; the first communication control component is provided with a first hose 7, and a first peristaltic pump 9 is installed on the first hose 7.

[0032] It should be noted that the first flexible hose 7 is made of polytetrafluoroethylene (PTFE), which is suitable for conveying corrosive media such as desulfurization slurry. The first peristaltic pump 9 has its operating time and speed controlled by the host computer 16. The two ends of the first flexible hose 7 are connected to the outlet of the slurry storage tank 12 and the inlet of the reactor 1, respectively.

[0033] It should be noted that the host computer 16 sends control commands to the first peristaltic pump 9 according to the detection requirements. The first peristaltic pump 9 drives the rotor of the stepper motor to squeeze the first hose 7, drawing the slurry from the slurry storage tank 12 into the hose and pushing it into the reactor 1. The volume of slurry delivered to the reactor 1 is controlled by adjusting the running time and speed of the peristaltic pump.

[0034] The water storage tank 13 is connected to the reactor 1 through the second communication control component; the second communication control component is provided with a second hose 8, and a second peristaltic pump 10 is installed on the second hose 8.

[0035] In some embodiments, the second peristaltic pump 10 is a variable metering peristaltic pump, controlled by a host computer 16. The two ends of the second hose 8 are connected to the outlet of the water storage tank 13 and the inlet of the reactor 1.

[0036] Optionally, a solenoid valve is installed between the outlet of the water storage tank 13 and the second peristaltic pump 10. The solenoid valve is linked to the liquid level switch 14 of the water storage tank 13. When the liquid level in the water storage tank is lower than the lower limit, the solenoid valve automatically closes to prevent the pump from running dry and damaging the equipment.

[0037] Optionally, the host computer 16 sends an operating command to the second peristaltic pump 10 according to the required dilution ratio for detection. The second peristaltic pump 10 squeezes the second hose 8 to generate suction, drawing in the dilution water in the water storage tank 13 and delivering it to the reactor 1.

[0038] By controlling the operating parameters of the peristaltic pump, the delivery volume of the dilution water is controlled, forming a preset dilution ratio with the slurry delivered by the first communication control component.

[0039] When the water level in the water tank 13 is lower than the preset lower limit, the level switch 14 sends a signal to the host computer, which then controls the solenoid valve to close and suspends the operation of the second peristaltic pump 10.

[0040] The chloride ion correction unit is connected to reactor 1; the chloride ion correction unit is equipped with a standard chloride ion solution storage tank 15, which is connected to reactor 1 through a third peristaltic pump 11.

[0041] In some embodiments, the third peristaltic pump 11 is a micro-metering peristaltic pump, which is connected to the tank outlet and the calibration liquid port of the reactor 1 at both ends of the connecting pipeline.

[0042] A solenoid valve is installed at the outlet of the standard solution storage tank 15, which is controlled by the host computer 16. A one-way valve is installed at the end of the pipeline between the third peristaltic pump 11 and the reactor 1 to prevent the liquid in the reactor from flowing back to the calibration unit.

[0043] It should be noted that when calibration is required, the host computer 16 controls the solenoid valve at the outlet of the standard solution storage tank 15 of the corresponding concentration to open according to the calibration requirements, and then sends an operation command to the third peristaltic pump 11. The third peristaltic pump 11 injects the standard chloride ion solution precisely into the reactor 1 by squeezing the tubing. After the signal of the chloride ion electrode 2 stabilizes, the calibration at that concentration is completed.

[0044] After calibration, the host computer shuts off the solenoid valve and the third peristaltic pump, and controls the sprayer 6 to flush the connecting pipes and reactor to prevent residual standard solution from affecting subsequent tests.

[0045] The reactor 1 is equipped with a chloride ion electrode 2, a temperature sensor 3, a double-layer stirrer 4, a pH meter 5, and a sprayer 6.

[0046] In some embodiments, the chloride ion electrode 2 is a solid membrane chloride ion selective electrode. The electrode probe is inserted into the reactor 1 at a predetermined depth and fixed by a fixing bracket 18. The bracket can adjust the electrode insertion depth.

[0047] Temperature sensor 3 uses a PT100 platinum resistance sensor, and the probe is installed adjacent to chloride ion electrode 2 to ensure that the temperature of the same reaction area is detected.

[0048] The double-layer mixer 4 is driven by a stepper motor and equipped with two layers of blades, the upper layer of blades being propeller-type and the lower layer of blades being turbine-type.

[0049] pH meter 5 can detect the pH value of the solution. The sensor probe in pH meter 5 is installed in the middle of the reactor.

[0050] The sprayer 6 is installed on the inner side of the top of the reactor and arranged around the fixed bracket 18. It has multiple spray nozzles and is connected to the high-pressure branch of the water storage tank 13, which can realize the all-round washing of the inner wall of the reactor, each sensor probe and the agitator blade.

[0051] As can be seen, the chloride ion electrode 2 detects the chloride ion concentration in the reactor solution in real time and outputs a potential signal; the temperature sensor 3 detects the solution temperature and provides data for temperature compensation of the chloride ion concentration measurement value.

[0052] The upper and lower blades of the double-layer agitator 4 rotate in tandem. The upper propulsion blades push the solution upward, while the lower turbine blades spread the solution outward, achieving rapid and uniform mixing of the slurry and dilution water.

[0053] pH meter 5 collects the pH value of the solution. After the test is completed, sprayer 6 sprays high-pressure atomized water to rinse the internal components and inner walls of the reactor and remove any residual solution.

[0054] The host computer 16 is electrically connected to the first communication control component, the second communication control component, the pH detector 5, the chloride ion electrode 2, the temperature sensor 3, the double-layer stirrer 4, and the sprayer 6. The host computer 16 is used to obtain the background concentration value C0 of the dilution water, and to obtain the mixing value C1 by controlling the pumping and mixing of the slurry and the dilution water. Based on the comparison result of C1 and the preset threshold, the final concentration is calculated, or C2 is obtained by starting flushing and adjusting the dilution ratio, and C3 is obtained by direct measurement of the slurry. Finally, the actual chloride ion concentration of the slurry is calculated using the effective measured value Cn, the background chloride ion concentration value C0, and the volume parameters.

[0055] In some embodiments, the host computer 16 controls the second peristaltic pump 10 to pump a quantitative amount of dilution water into the reactor 1, and after waiting for a set time, collects the measured value of the chloride ion electrode 2 as the background concentration value C0, and then controls the solenoid valve 17 to drain the dilution water.

[0056] The first peristaltic pump 9 and the second peristaltic pump 10 are controlled to pump the slurry and dilution water into the mixture according to a preset ratio. The double-layer agitator 4 is started to mix the mixture. After mixing is completed, C1 is collected. C1 is compared with the preset threshold Cs1. If C1 exceeds Cs1, it is considered valid and the actual concentration is calculated. If it does not exceed the threshold, the sprayer 6 is started to rinse the mixture. After adjusting the dilution ratio, the mixture is pumped back in to measure C2, which is then compared with Cs2.

[0057] If C2 is still ineffective, C3 is obtained by directly measuring the slurry after draining and flushing. Finally, based on the above effective measurement value Cn, combined with C0 and volume parameters, the actual chloride ion concentration of the slurry is calculated.

[0058] This embodiment enables the host computer to control each component, with adjustable concentration measurement waiting time to adapt to the stable response requirements of the chloride ion electrode under different operating conditions, thus improving measurement accuracy. Multiple calculations of volume parameters and actual concentrations reduce calculation errors.

[0059] In some embodiments, the host computer 16 combines a long short-term memory network to establish a multi-parameter correlation prediction model. It trains the model using historical detection data including chloride ion concentration, pH, and temperature, and predicts the chloride ion concentration in the future time period based on the model. According to the comparison between the predicted information and the actual working conditions, it controls the chloride ion correction unit, the first connection control component, the second connection control component, the double-layer stirrer 4, and the sprayer 6 to adjust the dilution ratio and stirring time, thereby realizing online monitoring of chloride ions.

[0060] The following are embodiments of the high-precision wide-range online monitoring method for chloride ions in slurry provided in this disclosure. This method belongs to the same inventive concept as the high-precision wide-range online monitoring system for chloride ions in slurry based on dynamic adaptation in the above embodiments. For details not described in detail in the embodiments of the high-precision wide-range online monitoring method for chloride ions in slurry, please refer to the embodiments of the high-precision wide-range online monitoring system for chloride ions in slurry based on dynamic adaptation described above.

[0061] S101: The host computer 16 monitors the liquid level of the water storage tank 13 and controls the water replenishment through the liquid level switch 14 to keep the liquid level of the water storage tank 13 at the set height.

[0062] In some embodiments, the level switch 14 floats up and down with the change of liquid level in the water tank 13, triggering different signal nodes to convert the liquid level information into an electrical signal and transmit it to the host computer 16.

[0063] The host computer 16 has a built-in liquid level judgment logic. It analyzes the received signal and determines whether the current liquid level is within the preset liquid level range. If it exceeds the range, it controls the start and stop of the water supply valve to achieve dynamic adjustment of the liquid level and maintain the stability of the water level in the storage tank.

[0064] S102: Control the second peristaltic pump 10 to pump a quantitative amount of dilution water into reactor 1, control the chloride ion electrode 2 to measure the background chloride ion concentration value C0, and control the solenoid valve 17 to empty reactor 1.

[0065] In some embodiments, the host computer 16 sends a control command to the second peristaltic pump 10 according to preset quantitative parameters to pump dilution water into the reactor 1. After the dilution water injection is completed, and the state of the dilution water in the reactor 1 is stable, the chloride ion electrode 2 is controlled to start measurement, continuously collect potential signals, take the average value and convert it into the background chloride ion concentration value C0 and store it.

[0066] The host computer 16 sends an opening command to the solenoid valve 17 of the drain pipe at the bottom of the reactor 1. After the solenoid valve 17 is energized, the valve core opens, and the dilution water is drained by the gravity of the conical bottom of the reactor 1, ensuring that there is no residue.

[0067] S103: Control the first peristaltic pump 9 to pump a quantitative amount of slurry and control the second peristaltic pump 10 to pump a quantitative amount of dilution water into the reactor 1. After starting the double-layer stirrer 4 to mix, control the chloride ion electrode 2 to measure the concentration value C1 of the diluted mixed chloride ions.

[0068] In some embodiments, the host computer 16 sends instructions to the first peristaltic pump 9 and the second peristaltic pump 10 to set the flow rate and running time of the first peristaltic pump 9 and the flow rate and running time of the second peristaltic pump 10, so that the slurry and dilution water are mixed in a preset ratio.

[0069] After the material is pumped in, the host computer 16 controls the double-layer agitator 4 to start. After running for a certain period of time, the upper and lower blades work together to achieve full mixing of the material.

[0070] After stirring stops, wait for the mixture to settle and stabilize, then control the chloride ion electrode 2 to start measurement, collect the potential signal over a certain period of time and take the average value, convert it into the diluted mixed chloride ion concentration value C1, and store it in association with the volume parameters pumped in this time.

[0071] S104: Determine whether C1 exceeds the set limit value Cs1: If C1 exceeds the limit value CS1, then C1 is determined to be valid, and step S108 is executed; if C1 does not exceed the limit value CS1, then C1 is determined to be invalid, and step S105 is executed.

[0072] In some embodiments, the limit value Cs1 is set according to the performance of the chloride ion electrode 2, and can be set according to the lowest effective measurement concentration of the electrode in a dilution environment, and stored in the memory of the host computer 16. The host computer 16 retrieves the stored C1 measurement value and compares it with Cs1.

[0073] If the measured value of C1 is greater than the set minimum effective measurement concentration, C1 is marked as valid in the system, and the C1 value is associated with the corresponding volume parameter and C0 value, and pushed to S108 to execute the calculation process.

[0074] If the measured value of C1 is less than or equal to the set minimum effective measurement concentration, C1 is marked as invalid. The volume ratio of slurry to dilution water in this S103 is recorded to provide a reference for the ratio adjustment in S105.

[0075] As can be seen, the Cs1 setting closely matches the actual electrode performance, ensuring the accuracy of the concentration results. Recording the current dilution ratio when ineffective allows for more targeted adjustments in subsequent steps, improving process efficiency.

[0076] S105: After controlling the solenoid valve 17 to drain and controlling the sprayer 6 to rinse the reactor 1, adjust the volume ratio of dilution water to slurry, re-pump the mixed liquid, and measure the concentration value C2 of the diluted mixed chloride ions.

[0077] In some embodiments, the host computer 16 controls the solenoid valve 17 to open, emptying the ineffective mixture in the reactor 1. After emptying, it sends a start command to the sprayer 6, which sprays high-pressure atomized water to rinse the inner wall of the reactor, the chloride ion electrode 2 probe, and the blades of the double-layer agitator 4. After rinsing is completed, the host computer 16 controls the solenoid valve 17 to open again, emptying the rinsing wastewater.

[0078] Based on the 1:3 dilution ratio recorded in S104, the host computer 16 adjusts the volume ratio. If the previous ratio was 1:3, it can be adjusted to 1:2. The host computer 16 sets the flow rate and running time of the first peristaltic pump 9 and the flow rate and running time of the second peristaltic pump 10 respectively. After controlling the two pumps to pump the material again, the double-layer agitator 4 is started. After stirring, the chloride ion electrode 2 is controlled to measure and store the C2 value.

[0079] High-pressure atomized rinsing removes residues, prevents cross-contamination between different batches of materials, and ensures a clean C2 measurement environment. Targeted adjustments to the dilution ratio ensure C2 remains within its effective measurement range, improving overall testing efficiency.

[0080] S106: Determine whether C2 exceeds the set limit value CS2: If C2 exceeds the limit value CS2, then C2 is determined to be valid, and C2 is used to replace C1. Proceed to step S108; if C2 does not exceed the limit value CS2, then C2 is determined to be invalid. Proceed to step S107.

[0081] In some embodiments, the threshold CS2 is higher than Cs1, and the setting is based on the premise that the concentration of the mixture should increase after adjusting the dilution ratio, and the measurement validity is further confirmed by a higher threshold.

[0082] The host computer 16 retrieves the measured value of C2 and compares it with CS2. If the measured value of C2 is greater than CS2, C2 is marked as valid, the original associated value of C1 is replaced with the value of C2, the corresponding volume parameters are updated, and the calculation process is pushed to S108. If the measured value of C2 is less than or equal to CS2, C2 is marked as invalid, and a direct measurement command is generated to prepare for the execution of step S107.

[0083] Optionally, CS2 is set based on the expected concentration after dilution ratio adjustment. A higher threshold is used to ensure that C2 not only falls within the effective measurement range of the electrode but also accurately reflects the actual concentration level of the slurry. If effective, it replaces C1 for subsequent calculations; if ineffective, it indicates that the dilution method can no longer meet the measurement requirements, and the system needs to switch to direct measurement mode.

[0084] S107: After controlling the solenoid valve 17 to drain and controlling the sprayer 6 to rinse the reactor 1, control the first peristaltic pump 9 to pump in a quantitative slurry for direct measurement, and measure the chloride ion concentration value C3. Use C3 as the final measurement value and execute step S108.

[0085] In some embodiments, the host computer 16 controls the solenoid valve 17 to open, emptying the ineffective mixture in the reactor 1, and then starts the sprayer 6 to rinse, extending the rinsing time by a certain amount compared to the previous rinsing time, to ensure that the residual diluent and slurry components are thoroughly removed.

[0086] After rinsing is completed, the solenoid valve 17 is opened to drain the rinsing water, and then a command is sent to the first peristaltic pump 9 to set the flow rate and duration, and the slurry is pumped into the reactor 1 without adding any dilution water.

[0087] After the slurry injection is completed, the chloride ion electrode 2 is controlled to start the measurement, collect the potential signal over a certain period of time and take the average value, convert it into a chloride ion concentration value C3, mark it as the final valid measurement value and store it.

[0088] S108: The host computer 16 calculates the actual chloride ion concentration of the slurry based on the effective measured value Cn, the background chloride ion concentration value C0, and the volume parameters. Cn represents the effective value of C1, C2, or C3.

[0089] In some embodiments, the host computer 16 first retrieves the corresponding valid measurement value Cn (C1, C2 or C3), the background chloride ion concentration value C0 measured in S102, and the volume parameter corresponding to Cn from the memory.

[0090] If Cn is C1 or C2, the host computer 16 retrieves the corresponding slurry volume V (slurry) and dilution water volume V (dilution water), and calculates the total volume V (total) = V (slurry) + V (dilution water); if Cn is C3, there is no need to retrieve the dilution water volume, V (total) = V (slurry), and the C0 value is not included in the calculation.

[0091] The built-in concentration inversion formula is used: C(actual chloride ion concentration) = [C1 × V(total) - C0 × V(dilution water)] / V(slurry). Where V(total) = V(dilution water) + V(slurry).

[0092] When C3 is reached, the formula simplifies to C (actual chloride ion concentration) = C3, thus completing the calculation.

[0093] After the calculation is completed, the actual concentration value is displayed in real time on the touch screen of the host computer 16 and stored.

[0094] As can be seen, the background chloride ion concentration value C0 is used to subtract the influence of chloride ions contained in the dilution water on the concentration of the mixed liquor. The volume parameter provides the material ratio basis for concentration inversion. The actual chloride ion concentration of the original slurry can be deduced from the mixed liquor concentration or the directly measured concentration through formula calculation. The formula adaptation logic corresponding to different Cn values ​​is displayed and stored in real time, making it convenient for operators to view the test results.

[0095] In one embodiment of the present invention, the following steps are included after S108. S201: The host computer 16 collects the chloride ion concentration, pH value and temperature value output by the chloride ion electrode 2, pH detector 5 and temperature sensor 3 during the historical period, and simultaneously records the running time of the first peristaltic pump 9, the running time of the second peristaltic pump 10 and the running time of the double-layer stirrer 4 at the corresponding time as operation tags.

[0096] S202: The host computer 16 performs outlier removal and missing value interpolation on the collected historical data, normalizes each physical quantity to the [0,1] interval, and extracts the rate of change, local extrema and mean within the sliding time window as feature vectors.

[0097] In some embodiments, field data often exhibits anomalous jumps or gaps due to sensor drift, communication interruptions, or bubble interference. The host computer 16 identifies points deviating from the mean by more than three standard deviations as outliers and removes them. For segments with no more than three consecutive missing sampling points, linear interpolation is used to fill the gaps. Longer missing segments are marked as invalid periods and excluded from training.

[0098] Chloride ion concentration, pH, and temperature are linearly compressed to the [0,1] interval according to their respective ranges to eliminate dimensional differences.

[0099] Optionally, for each parameter, the average, maximum, minimum, and first-difference mean over the past 5, 15, and 30-minute windows are calculated to form a multi-scale time feature. This processing enables the model to capture short-term fluctuations and long-term trends, enhancing its sensitivity to concentration abrupt changes.

[0100] S203: The host computer 16 trains a multi-parameter correlation prediction model based on the processed feature vector. The model takes the multi-parameter features of the current and previous N sampling periods as input and outputs the recommended dilution water volume, slurry volume and double-layer agitator 4 running time for the next detection period.

[0101] In some embodiments, the host computer 16 is configured with a Long Short-Term Memory (LSTM) network. The input layer receives multi-parameter, multi-scale feature vectors, the hidden layer contains 128 memory units, and the output layer contains three neurons, corresponding to the recommended dilution water volume, slurry volume, and stirring time, respectively.

[0102] Optionally, the volume of dilution water is converted from the running time of the second peristaltic pump 10, and the volume of slurry is converted from the running time of the first peristaltic pump 9.

[0103] The model employs a mean squared error loss function and is pre-trained offline on historical datasets until the validation set error converges. The model can memorize concentration evolution patterns over the past few hours. For example, when the load of the desulfurization system increases, it is often accompanied by a slow accumulation of chloride ions. The model can increase the sampling frequency or reduce the dilution ratio in advance to prevent entering direct measurement mode.

[0104] S204: Before executing step S103, the host computer 16 calls the trained multi-parameter correlation prediction model to generate the dilution water injection volume, slurry injection volume and stirring time required for this test based on the latest collected chloride ion concentration trend, pH value and temperature value, and sets the start and stop duration of the first peristaltic pump 9 and the second peristaltic pump 10 and the running time of the double-layer agitator 4 accordingly.

[0105] In some embodiments, whenever the system is about to enter a new detection cycle, that is, after S102 has completed the measurement of the background chloride ion concentration value C0 and before S103 has been executed, the host computer 16 reads the pH, temperature and chloride ion concentration trend slope of the most recent slurry state in the external pipeline of the current reactor 1, splices them into a real-time feature vector, and inputs it into the trained multi-parameter correlation prediction model.

[0106] The three values ​​output by the model are converted into the target running seconds of the first peristaltic pump 9 and the second peristaltic pump 10, and the target running seconds of the double-layer agitator 4 after inverse normalization.

[0107] The host computer 16 sets the pump and valve operating parameters accordingly, replacing the fixed dilution ratio in the original scheme. This method enables the system to automatically use a low dilution ratio during the high concentration rise period and a high dilution ratio during the low concentration stabilization period, ensuring that the electrode operates in the linear region.

[0108] S205: After completing the calculation of the actual chloride ion concentration in step S108, the host computer 16 compares the measured effective concentration value Cn with the concentration value previously predicted by the model, calculates the prediction error, and uses the error to fine-tune the weight parameters of the multi-parameter correlation prediction model online through the backpropagation algorithm.

[0109] In some embodiments, after each successful calculation of the actual chloride ion concentration by S108, the host computer 16 retrieves the concentration value predicted by the multi-parameter correlation prediction model before the start of this detection, and calculates the absolute error with the measured value Cn.

[0110] If the error exceeds the preset threshold, online learning is triggered. The complete input feature vector and the true output, i.e. the dilution ratio and stirring time corresponding to the effective Cn, are combined to form a new sample. The stochastic gradient descent method is used to fine-tune the weights of the multi-parameter correlation prediction model in a single step.

[0111] The weight update magnitude is limited by the learning rate to prevent noisy data from causing model oscillations. This allows it to adapt to slow changes in slurry composition and maintain prediction accuracy.

[0112] As one implementation of the present invention, the following steps are included after step S205: S206: The host computer 16 extracts the predicted chloride ion concentration value Cp output by the model before the start of this detection cycle from the internal cache of the multi-parameter correlation prediction model. Cp is derived by the model based on the input feature vector.

[0113] In some embodiments, the multi-parameter correlation prediction model calculates an implicit chloride ion concentration prediction value Cp each time it is invoked to generate the dilution ratio and stirring time. Cp is not used directly for control, but rather serves as a basis for consistency verification of the model's internal state.

[0114] After the model inference is completed, the host computer 16 temporarily stores Cp to ensure that the subsequent error calculation uses the prediction result that is completely matched with the current operation.

[0115] S207: The host computer 16 calculates the difference between the measured effective concentration value Cn and the predicted concentration value Cp to obtain the absolute error ΔC = |Cn|. Cp|, and determine whether ΔC is greater than the preset error threshold ε.

[0116] In some embodiments, the host computer 16 calculates the absolute deviation ΔC between the measured value Cn and the predicted value Cp, and compares it with a preset error threshold ε. When the deviation exceeds the threshold, the model is considered to have a systematic deviation and needs to be corrected. If the deviation is within the tolerance, the subsequent update process is skipped to prevent noisy data from causing model oscillations.

[0117] The error threshold ε can be dynamically adjusted according to different operating conditions. For example, ε can be reduced during the rapid increase of concentration to improve sensitivity.

[0118] S208: If ΔC>ε, then the host computer 16 will combine the input feature vector X corresponding to this detection, the real operation parameters Y including the actual volume of dilution water, the volume of slurry, the stirring time, and the target output label T, i.e., Cn, to deduce the ideal operation parameters required to back-calculate the target output label T, i.e., Cn, into a new training sample.

[0119] In some embodiments, the actual operating parameters Y include the actual volume of dilution water, slurry volume, and stirring time. The target output label T is the ideal operating parameters required to back-calculate Cn.

[0120] The actual operating parameter Y here is not the model recommended value, but the actual action performed, such as the duration of operation of the first peristaltic pump 9, which corresponds to the slurry volume. The target output label is the ideal dilution ratio and stirring time combination derived from the relationship between Cn, C0 and volume.

[0121] This construction method enables the model to learn what actions should be taken under given operating conditions to obtain accurate measurements. The samples contain multidimensional inputs and desired outputs, forming the basic unit of supervised learning.

[0122] S209: The host computer 16 adds the new training samples to the online fine-tuning queue and performs a single forward and backward propagation update on the weight matrix of the multi-parameter correlation prediction model using the mean squared error as the loss function.

[0123] In some embodiments, the host computer 16 does not retrain the entire multi-parameter association prediction model, but performs a forward calculation on the latest sample to obtain the loss value, then calculates the gradient of each weight through backpropagation, and updates the weight matrix with a preset learning rate.

[0124] S210: After the weight update is completed, the host computer 16 stores the updated multi-parameter correlation prediction model parameters and releases the cached prediction value Cp, feature vector X and error ΔC to prepare the running space for the next detection cycle.

[0125] In some embodiments, after the weights are updated, the host computer 16 stores the new parameters and clears the temporarily cached data in the current detection cycle, including feature vector X, predicted value Cp, error ΔC, etc.

[0126] In one embodiment of the present invention, the training process of the multi-parameter correlation prediction model for online monitoring of chloride ions includes the following steps: S301: The host computer 16 collects data from various sensors and actuators, including readings from the chloride ion electrode 2, pH meter 5, temperature sensor 3, and operation records of the first peristaltic pump 9, the second peristaltic pump 10, and the double-layer stirrer 4, to construct the original training set.

[0127] S302: Preprocess the original training set, including outlier removal, missing value imputation, and normalization. Statistical methods are used to identify and remove data points that deviate from the mean by more than three standard deviations; for missing values, data from adjacent time points are used to fill them using linear interpolation. All physical quantities are scaled to the [0,1] interval to ensure that data of different dimensions can be compared and analyzed on the same scale.

[0128] S303: Within each sampling period, calculate the statistical characteristics such as the average, maximum and minimum values, and rate of change of the sliding window over a past period to form a feature vector set.

[0129] For example, the above statistical characteristics are calculated by selecting time windows of the past 5 minutes, 15 minutes, and 30 minutes respectively, in order to capture short-term fluctuations and long-term trends.

[0130] In some embodiments, meaningful information is extracted from the raw data. Specific operations include, but are not limited to, calculating averages, extreme values, and rates of change over historical time periods.

[0131] This method can reflect the current state and reveal potential trends and patterns. Through feature extraction, the model can understand the patterns in the input data, thereby making accurate predictions.

[0132] S304: Based on the feature vector set, construct and train a multi-parameter association prediction model based on the Long Short-Term Memory (LSTM) network. The multi-parameter association prediction model consists of an input layer, a hidden layer with 128 memory units, and an output layer.

[0133] The input layer receives feature vectors, the hidden layer is responsible for learning the correlation between sequences, and the output layer provides recommended values ​​for dilution water volume, slurry volume, and stirring time.

[0134] The mean squared error is used as the loss function, and the weights are adjusted using the batch stochastic gradient descent algorithm until the model converges.

[0135] In some embodiments, the LSTM model is chosen because it excels at handling time-sequential data, making it suitable for the needs of this application scenario. Model complexity and computational efficiency are balanced by adjusting the number of hidden layers and memory units.

[0136] The training process utilizes historical data to optimize model parameters, enabling accurate prediction of future behavior given input conditions. This deep learning-based approach can uncover complex nonlinear relationships within data.

[0137] S305: After initial training, online updates are performed using newly acquired data. After each actual measurement, the effective concentration value Cn is compared with the previously predicted value, and the model weights are adjusted based on the error.

[0138] In this way, the multi-parameter correlation prediction model can adapt to changes in the system over time and maintain high prediction accuracy.

[0139] In some embodiments, after each new detection period, the multi-parameter association prediction model is updated based on the latest measured results. If there is a significant difference between the predicted and actual values, it indicates that the multi-parameter association prediction model needs to be adjusted.

[0140] By using the backpropagation algorithm, the model weights are fine-tuned based on the latest error, enabling the multi-parameter correlation prediction model to gradually adapt to changes in the system.

[0141] This approach ensures that the multi-parameter correlation prediction model maintains good predictive performance even when faced with changes in external conditions.

[0142] In one embodiment of the present invention, based on step S104, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S104 specifically includes the following steps: S1041: The fluctuation range of chloride ion concentration in the historical detection data of the host computer 16 and the influence of the historical temperature information fed back by the temperature sensor 3 on the chloride ion concentration measurement, and set the specific value of the limit value CS1.

[0143] This embodiment analyzes the variation of chloride ion concentration measurements at different temperatures to ensure that CS1 is higher than the lowest chloride ion concentration that may occur after normal dilution of the slurry within a common temperature range, and lower than the concentration value corresponding to the maximum allowable measurement error range of the system.

[0144] In some embodiments, temperature has a certain impact on the measurement of chloride ion concentration. The measurement characteristics of chloride ion electrode 2 may change at different temperatures, resulting in deviations in the measured values.

[0145] The host computer 16 combines historical temperature information to analyze the relationship between temperature and chloride ion concentration measurement. When setting CS1, it takes temperature factors into account, which makes the setting of CS1 more in line with the actual working conditions.

[0146] In this way, the validity of the measured value can be determined. For example, at higher temperatures, the measured value of chloride ion concentration may change relative to the normal temperature. By taking temperature factors into account when setting CS1, the judgment standard can be guaranteed to be consistent with the actual situation.

[0147] S1042: The host computer 16 obtains the diluted mixed chloride ion concentration value C1 obtained from the chloride ion electrode 2, and at the same time reads the temperature value T currently measured by the temperature sensor 3 and records the corresponding numerical data.

[0148] S1043: The host computer 16 corrects the obtained C1 value according to the preset temperature and concentration correction relationship and the current temperature value T, to obtain the corrected concentration value C1c. C1c is compared with the set CS1 value, and internal logic operations are used to determine whether C1c is greater than CS1.

[0149] In some embodiments, since temperature affects the measurement accuracy of chloride ion electrode 2, the measured value can be corrected by a pre-set correction relationship to obtain a value closer to the true concentration.

[0150] This embodiment compares the corrected concentration value C1c with CS1 to determine whether the measured value meets the valid conditions, thus improving the accuracy of the judgment. For example, if the temperature causes the measured value to be too high, the correction can restore a more realistic concentration situation.

[0151] S1044: If the comparison result is that C1c is greater than CS1, the host computer 16 will determine that C1 is valid and trigger the instruction to execute step S108; if the comparison result is that C1c is less than or equal to CS1, then C1 is determined to be invalid and the instruction to execute step S105 will be triggered.

[0152] In some embodiments, when the corrected concentration value C1c is valid, it indicates that the current measured value is within a reasonable range, and the concentration can be directly calculated, thereby improving detection efficiency.

[0153] When the result is invalid, the dilution ratio is adjusted and the measurement is repeated to ensure an accurate concentration value is obtained. This process branch control based on the correction result allows the system to adapt to the effects of temperature changes, ensuring the reliability of the test results.

[0154] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0155] like Figure 3 As shown, this application also provides a host computer, including a display module 103, a memory 102, a processor 101, a communication module 104, and a computer program stored in the memory and executable on the processor 101. When the processor 101 executes the program, it implements the steps of a high-precision, wide-range online monitoring method for chloride ions in slurry.

[0156] In embodiments of the present invention, the host computer includes, but is not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The host computer can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the embodiments described and / or claimed herein.

[0157] In this embodiment, processor 101 may be implemented using at least one of an application-specific integrated circuit, a programmable logic device, a field-programmable gate array, a processor, a controller, a microcontroller, a microprocessor, or an electronic unit designed to perform the functions described herein. In some cases, such an implementation may be implemented within a controller. For software implementation, implementations such as processes or functions may be implemented with separate software modules that allow the performance of at least one function or operation. Software code may be implemented by a software application (or program) written in any suitable programming language, and the software code may be stored in memory and executed by the controller.

[0158] The display module 103 is used to display information input by the user or information provided to the user. The display module 103 may include a display panel, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like.

[0159] The memory 102 can be used to store software programs and various data. The memory 102 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0160] The communication module 104 transmits radio signals to and / or receives radio signals from at least one of a base station, an external terminal, and a server. Such radio signals may include voice call signals, video call signals, or various types of data sent and / or received according to text and / or multimedia messages.

[0161] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A high-precision, wide-range online monitoring system for chloride ions in slurry based on dynamic adaptive principles, characterized in that, include: Reactor (1), slurry storage tank (12), water storage tank (13) and host computer (16); The slurry storage tank (12) is connected to the reactor (1) through the first communication control component; The water storage tank (13) is connected to the reactor (1) via the second communication control component; The reactor (1) is equipped with a chloride ion electrode (2), a temperature sensor (3), a double-layer stirrer (4), a pH meter (5), and a sprayer (6). The host computer (16) is electrically connected to the first communication control component, the second communication control component, the pH detector (5), the chloride ion electrode (2), the temperature sensor (3), the double-layer stirrer (4), and the sprayer (6), respectively.

2. The high-precision, wide-range online monitoring system for chloride ions in slurry based on dynamic adaptation according to claim 1, characterized in that, Also includes: Chloride ion correction unit; The chloride ion correction unit is connected to the reactor (1); The chloride ion correction unit is equipped with a standard chloride ion solution storage tank (15), which is connected to the reactor (1) via a third peristaltic pump (11); The first communication control component is provided with a first hose (7), and a first peristaltic pump (9) is installed on the first hose (7). The second communication control component is provided with a second hose (8), and a second peristaltic pump (10) is installed on the second hose (8).

3. The high-precision, wide-range online monitoring system for chloride ions in slurry based on dynamic adaptation according to claim 1, characterized in that, A level switch (14) is installed on the side wall of the water storage tank (13); A drain pipe is provided at the bottom of the reactor (1), and a solenoid valve (17) is installed on the drain pipe. A fixed bracket (18) is installed near the top of the reactor (1), and the chloride ion electrode (2), temperature sensor (3), double-layer stirrer (4), pH meter (5) and sprayer (6) are fixed on the fixed bracket (18).

4. The high-precision, wide-range online monitoring system for chloride ions in slurry based on dynamic adaptive principle according to claim 1, characterized in that, The host computer (16) combines a long short-term memory network to establish a multi-parameter correlation prediction model. It uses historical detection data including chloride ion concentration, pH, and temperature to train the model and predicts the chloride ion concentration in the future time period based on the model. According to the comparison between the prediction information and the actual working conditions, it controls the chloride ion correction unit, the first connection control component, the second connection control component, the double-layer stirrer (4), and the sprayer (6) to adjust the dilution ratio and stirring time, so as to realize online monitoring of chloride ions.

5. A high-precision, wide-range online monitoring method for chloride ions in slurry, characterized in that, The method is implemented based on the high-precision, wide-range online monitoring system for slurry chloride ions based on dynamic adaptation as described in any one of claims 1 to 4, and the method includes: S101: The host computer (16) monitors the liquid level of the water storage tank (13) and controls the water replenishment through the liquid level switch (14) to keep the liquid level of the water storage tank (13) at the set height; S102: Control the second peristaltic pump (10) to pump a quantitative amount of dilution water into the reactor (1), control the chloride ion electrode (2) to measure the background chloride ion concentration value C0, and control the solenoid valve (17) to empty the reactor (1). S103: Control the first peristaltic pump (9) to pump a quantitative amount of slurry and control the second peristaltic pump (10) to pump a quantitative amount of dilution water into the reactor (1). Start the double-layer stirrer (4) to mix, and control the chloride ion electrode (2) to measure the concentration value C1 of the diluted mixed chloride ion. S104: Determine whether C1 exceeds the set limit value Cs1: If C1 exceeds the limit value CS1, then C1 is determined to be valid, and step S108 is executed; If C1 does not exceed the limit value CS1, then C1 is determined to be invalid, and step S105 is executed; S105: After controlling the solenoid valve (17) to drain and controlling the sprayer (6) to rinse the reactor (1), adjust the volume ratio of dilution water to slurry, re-pump the mixture, and measure the concentration value of chloride ions C2 after dilution. S106: Determine whether C2 exceeds the set limit value CS2: If C2 exceeds the limit value CS2, then C2 is determined to be valid, and C2 is used to replace C1, and step S108 is executed; If C2 does not exceed the limit value CS2, then C2 is determined to be invalid, and step S107 is executed; S107: After controlling the solenoid valve (17) to drain and controlling the sprayer (6) to rinse the reactor (1), control the first peristaltic pump (9) to pump in a quantitative slurry for direct measurement, and measure the chloride ion concentration value C3. Use C3 as the final measurement value and execute step S108. S108: The host computer (16) calculates the actual chloride ion concentration of the slurry based on the effective measured value Cn, the background chloride ion concentration value C0 and the volume parameters. Cn represents the effective value of C1, C2 or C3.

6. The high-precision, wide-range online monitoring method for chloride ions in slurry according to claim 5, characterized in that, In S108, the method for calculating the actual chloride ion concentration in the slurry is as follows: C (actual chloride ion concentration) = [C1 × V (total) - C0 × V (dilution water)] / V (slurry) Where V(total) = V(dilution water) + V(slurry).

7. The high-precision, wide-range online monitoring method for chloride ions in slurry according to claim 5, characterized in that, Following S108 are: S201: The host computer (16) collects the chloride ion concentration, pH value and temperature value output by the chloride ion electrode (2), pH detector (5) and temperature sensor (3) during the historical period, and simultaneously records the running time of the first peristaltic pump (9), the running time of the second peristaltic pump (10) and the running time of the double-layer stirrer (4) at the corresponding time as operation tags; S202: Perform outlier removal and missing value imputation on the collected historical data, normalize each physical quantity to the [0,1] interval, and extract the rate of change, local extrema and mean within the sliding time window as feature vectors; S203: Train a multi-parameter correlation prediction model based on the processed feature vectors, and take the multi-parameter features of the current and previous N sampling periods as inputs to output the recommended dilution water volume, slurry volume and double-layer agitator (4) running time for the next detection period; S204: Before executing step S102, the trained multi-parameter correlation prediction model is called to generate the dilution water injection volume, slurry injection volume and stirring time required for this test based on the latest collected chloride ion concentration trend, pH value and temperature value, and the start and stop time of the first peristaltic pump (9) and the second peristaltic pump (10) and the running time of the double-layer agitator (4) are set accordingly. S205: After completing the calculation of the actual chloride ion concentration, the effective concentration value Cn measured in this study is compared with the concentration value predicted by the model in the past, the prediction error is calculated, and the weight parameters of the multi-parameter correlation prediction model are fine-tuned using the error.

8. The high-precision, wide-range online monitoring method for chloride ions in slurry according to claim 5, characterized in that, The training process of a multi-parameter association prediction model includes the following steps: The host computer (16) acquires data from each sensor and constructs the original training set; Preprocess the original training set; Within each sampling period, calculate the statistical characteristics of the sliding window over a past period, such as the average, maximum, minimum, and rate of change, to form a feature vector set. Based on the feature vector set, a multi-parameter association prediction model based on a long short-term memory network is constructed and trained; the multi-parameter association prediction model includes an input layer, a hidden layer and an output layer. The input layer receives feature vectors, the hidden layer is responsible for learning the correlation between sequences, and the output layer provides recommended values ​​for dilution water volume, slurry volume, and stirring time. Mean squared error is used as the loss function, and the weights are adjusted through batch stochastic gradient descent until the model converges. After initial training, the model is updated online using newly collected data. After each actual measurement, the effective concentration value Cn is compared with the previously predicted value, and the model weights are adjusted according to the error.

9. The high-precision, wide-range online monitoring method for chloride ions in slurry according to claim 5, characterized in that, Following S205 are the following steps: S206: The host computer (16) extracts the predicted chloride ion concentration value Cp from the multi-parameter correlation prediction model before the start of this detection cycle; S207: Subtract the measured effective concentration value Cn from the predicted concentration value Cp to obtain the absolute error ΔC = |Cn|. Cp|, and determine whether ΔC is greater than the preset error threshold ε; S208: If ΔC>ε, then the complete input feature vector X, the true operation parameters Y, and the target output label corresponding to this detection will be combined to form a new training sample; S209: Add new training samples to the online fine-tuning queue, and perform a single forward and backward propagation update on the weight matrix of the multi-parameter association prediction model using mean squared error as the loss function; S210: After the weights are updated, the updated model parameters are stored, and the cached predicted values ​​Cp, feature vectors X, and errors ΔC are released to prepare the running space for the next detection cycle.

10. The high-precision, wide-range online monitoring method for chloride ions in slurry according to claim 5, characterized in that, S104 specifically includes the following steps: The fluctuation range of chloride ion concentration in the historical detection data of the host computer (16) and the influence of the historical temperature information fed back by the temperature sensor (3) on the chloride ion concentration measurement are set to the specific value of the limit value CS1. The diluted mixed chloride ion concentration value C1 obtained from the chloride ion electrode (2) is obtained in this measurement. At the same time, the temperature value T obtained from the temperature sensor (3) is read and the corresponding numerical data is recorded. Based on the pre-set temperature and concentration correction relationship, the obtained C1 value is corrected according to the current temperature value T to obtain the corrected concentration value C1c; The value of C1c is compared with the set value of CS1, and the internal logic operation is used to determine whether C1c is greater than CS1. If the comparison result is that C1c is greater than CS1, C1 will be determined to be valid and the instruction to execute step S108 will be triggered; if the comparison result is that C1c is less than or equal to CS1, C1 will be determined to be invalid and the instruction to execute step S105 will be triggered.

Images