A temperature control parameter generation method based on thermal inertia characteristics

Abstract

The present application relates to the technical field of electrolytic cell temperature control, and particularly relates to a temperature control parameter generation method based on thermal inertia characteristics, comprising: extracting response delay and amplitude attenuation through local thermal excitation, distinguishing temperature measurement installation delay from electrolytic cell body thermal inertia; determining installation contact state through same temperature and same flow reference comparison, improving matching accuracy of installation temperature measurement model; identifying cell body-electrolyte thermal inertia through inverse pipe side temperature and combining net heat input, reducing interference of sensor contact change on thermal response parameters; screening temperature control parameters through series simulation, taking into account temperature deviation, temperature fluctuation and heat exchange regulation stability, so as to reduce temperature overshoot under load fluctuation and frequent valve action, and improve accuracy and adaptability of electrolytic cell temperature control parameter generation.

Description

Technical Field

[0001] This invention relates to the field of electrolytic cell temperature control technology, and in particular to a method for generating temperature control parameters based on thermal inertia characteristics. Background Technology

[0002] In megawatt-level alkaline water electrolysis hydrogen production systems powered by fluctuating renewable energy sources, the DC current of the electrolyzer rises and falls with changes in wind and solar power output. This leads to variations in the electrochemical heat generated in the tank, the heat carried by the circulating electrolyte, and the heat removed from the heat exchanger. The electrolyte outlet temperature exhibits significant thermal inertia and transmission delay. During initial equipment commissioning, replacement or reinstallation of the outlet temperature sensor, maintenance of the outlet pipeline insulation layer, and seasonal changes in cooling conditions, the tightness between the temperature sensor and the pipeline wall, the thickness of the thermal interface, and heat dissipation conditions may change, resulting in additional delays and amplitude attenuation in the sensor's temperature response. Existing temperature controller parameters are typically manually or model-tuned based on sensor temperature changes or the overall temperature response of the electrolyzer. This makes it difficult to distinguish between the response delay caused by the temperature sensor installation interface and the thermal inertia of the tank-electrolyte body. Increased installation contact thermal resistance can easily be misinterpreted as an increase in the electrolyzer's time constant, leading to a mismatch between the proportional coefficient, integral time, and derivative time and the actual thermal process. This results in outlet temperature overshoot, prolonged settling time, or frequent reverse actions of the heat exchange regulating valve during load increases and decreases.

[0003] For example, CN114967782A discloses a method and system for controlling the operating temperature of an electrolytic cell based on heat balance. This method obtains the actual inlet temperature of the electrolytic cell and calculates the inlet temperature setpoint based on the outlet temperature setpoint, the net heat power of the electrolytic cell, the electrolyte volume flow rate, the electrolyte density, and the electrolyte specific heat capacity. Then, it adjusts the refrigerant flow controller according to the deviation between the actual inlet temperature and the inlet temperature setpoint to change the electrolyte inlet temperature. However, this method uses the output of the temperature acquisition device as the direct basis for heat balance and temperature control, without identifying the installation contact state of the temperature sensor, nor separating the installation contact thermal resistance, sensor heat capacity, and the thermal inertia of the cell-electrolyte. When the temperature sensor is reinstalled, the pipeline insulation structure is changed, or the thermal interface degrades, the additional temperature measurement delay will still be included in the inlet temperature deviation calculation. Therefore, it is difficult to accurately generate temperature controller parameters that match the thermal response of the electrolytic cell.

[0004] For example, CN113930805A discloses a temperature prediction and control method and device for an electrolytic hydrogen production system. This method predicts the inlet temperature, outlet temperature, and cooling water temperature based on the future current signal and the electrolyzer energy equation, heat transfer equation, and cooling water flow equation, and determines the cooling water valve opening at the next control moment under the constraints of temperature status and cooling water valve opening. However, the control object of this method is mainly the cooling water valve opening at the next moment, and it does not form a temperature control parameter generation process composed of proportional coefficient, integral time, and derivative time. Furthermore, its temperature prediction model does not identify the sensor installation contact thermal resistance and sensor's own thermal capacity as independent temperature measurement links. When the temperature measurement delay changes with the installation status, the thermal response delay in the prediction model may simultaneously include the measurement chain delay and the electrolyzer body delay, making it difficult to accurately extract the thermal inertia characteristics of the tank-electrolyte.

[0005] Given the problems of existing electrolytic cell temperature control technology, such as the lack of inclusion of the temperature measurement installation status in the temperature control parameter tuning process, the mixing of thermal delay in the measurement chain and the thermal inertia of the tank-electrolyte body, and the difficulty in adapting temperature control parameters to sensor reinstallation and changes in cooling conditions, this invention proposes a temperature control parameter generation method based on thermal inertia characteristics. This method determines the contact response characteristics of the temperature sensor through local thermal excitation independent of the circulating heat exchange path, establishes an installation-based temperature measurement heat transfer model based on the installation contact status, and inversely calculates the pipeline-side temperature response from this model. Furthermore, it identifies the tank-electrolyte thermal inertia characteristics based on the electrolytic cell's net heat input and the pipeline-side temperature response. Candidate temperature control parameter combinations are then loaded into a simulation model that includes the heat exchange execution process, the electrolytic cell thermal response process, and the installation-based temperature measurement heat transfer process for screening. This addresses the problem of mismatch between temperature control parameters and the actual electrolytic cell thermal response caused by changes in the temperature measurement installation status. Summary of the Invention

[0006] The purpose of this section is to outline some aspects of the embodiments of the present invention and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this section, the abstract and title of the invention. Such simplifications or omissions shall not be used to limit the scope of the present invention.

[0007] In view of the aforementioned existing problems, the present invention is proposed.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A method for generating temperature control parameters based on thermal inertia characteristics includes: applying local thermal excitation independent of the circulating heat exchange path to an electrolyte outlet temperature sensor, and forming contact response characteristics based on response delay and normalized amplitude attenuation. The contact response characteristics are compared with the reference contact response characteristics at the same temperature and flow rate to determine the installation contact state and select the installation temperature measurement heat transfer model. The sensor temperature response when the net heat input changes is inverted using the aforementioned installation temperature measurement heat transfer model to obtain the pipeline-side temperature response; the electrolytic cell thermal response model is identified based on the net heat input and pipeline-side temperature response, and the thermal inertia characteristics of the cell-electrolyte are extracted. The candidate temperature control parameter combinations are loaded into a simulation model that connects the heat exchange execution model, the electrolytic cell thermal response model, and the installation temperature measurement heat transfer model. The combination with the smallest temperature deviation is selected from the combinations that meet the temperature and heat exchange fluctuation constraints.

[0009] The beneficial effects of this invention are as follows: This invention extracts response delay and amplitude attenuation through local thermal excitation, distinguishing between temperature measurement installation delay and the thermal inertia of the electrolytic cell body; it determines the installation contact state by comparing with benchmarks of the same temperature and flow rate, improving the matching accuracy of the installation temperature measurement model; it identifies the thermal inertia of the cell body-electrolyte by inverting the pipeline side temperature and combining it with net heat input, reducing the interference of sensor contact changes on thermal response parameters; and it filters temperature control parameters through series simulation, taking into account temperature deviation, temperature fluctuation, and heat transfer regulation stability, thereby reducing temperature overshoot and frequent valve operation under load fluctuations, and improving the accuracy and adaptability of generating electrolytic cell temperature control parameters. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the 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. Wherein: Figure 1 This is a schematic flowchart of the temperature control parameter generation method based on thermal inertia characteristics as shown in this invention. Detailed Implementation

[0011] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0012] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort should fall within the scope of protection of this invention.

[0013] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0014] The following describes in detail a specific implementation method for generating temperature control parameters based on thermal inertia characteristics, using the application scenario of a megawatt-level alkaline water electrolysis hydrogen production system powered by renewable energy fluctuations. The implementation methods are illustrated using scenarios such as initial equipment commissioning, replacement or reinstallation of the outlet temperature sensor, maintenance of pipeline insulation, and seasonal changes in cooling conditions. The controlled objects include the alkaline electrolyzer, electrolyte circulation pipeline, electrolyte heat exchanger, heat exchange regulating valve, and outlet temperature sensor. The physical states of the controlled objects include the electrolyzer DC current, total electrolyzer voltage, circulating electrolyte mass flow rate, electrolyte inlet temperature of the electrolyte heat exchanger, electrolyte outlet temperature of the electrolyte heat exchanger, cooling water inlet temperature, ambient temperature, and outlet temperature sensor temperature. The execution responses include the heat exchange regulating valve opening response, heat exchange removal response, and outlet temperature response. The physical states are collected by a DC current transmitter, DC voltage transmitter, Coriolis mass flow meter, platinum resistance temperature sensor, ambient temperature sensor, and valve position transmitter, and transmitted to the control station, which then forms a sampling sequence according to a unified time stamp.

[0015] In a preferred embodiment, the alkaline water electrolysis hydrogen production system has a rated input power of 5MW. The electrolyzer comprises 360 series-connected electrolysis chambers. The DC current operating range of the electrolyzer is 4200A to 6500A, the total voltage operating range is 680V to 780V, the circulating electrolyte mass flow rate operating range is 28kg / s to 36kg / s, the target outlet temperature is 80.0℃, and the upper limit of the electrolyzer operating temperature is 85.0℃. The outlet temperature sensor is installed inside the temperature measuring sleeve of the outlet circulation pipeline of the electrolyzer. The circulation pipeline uses 316L stainless steel pipes, and the space between the outlet temperature sensor and the temperature measuring sleeve is filled with a heat-conducting medium. The measurement resolution of the outlet temperature sensor is 0.01℃. The control station forms a unified sampling time according to a 1s sampling interval. The DC current of the electrolytic cell, the total voltage of the electrolytic cell, the voltage of the resistance heating element, and the current of the resistance heating element are obtained according to a 100ms acquisition period, and the arithmetic mean within each 1s sampling segment is calculated. The mass flow rate of the circulating electrolyte, the opening degree of the heat exchange regulating valve, and the measured values ​​of each temperature are obtained according to a 1s sampling interval.

[0016] According to an embodiment of the present invention, in combination Figure 1 The flowchart shown illustrates a method for generating temperature control parameters based on thermal inertia characteristics, which specifically includes the following steps: S1. Apply a localized thermal excitation independent of the circulating heat exchange path to the electrolyte outlet temperature sensor, and form contact response characteristics based on response delay and normalized amplitude attenuation. Note that the following should be noted in this step: S1.1 When the DC current of the electrolytic cell, the flow rate of the circulating electrolyte, and the opening degree of the heat exchange regulating valve are constant, and the reference temperature on the pipeline side is in a stable range, a pulse power with the same amplitude and duration as the reference contact sample is applied to the resistance heating element attached to the pipeline side of the temperature sensor mounting interface; the pipeline side reference temperature is determined according to the ratio of voltage to current across the resistance heating element and the resistance temperature characteristics, forming a pipeline side reference temperature response, and a sensor temperature response is formed through the temperature sensor.

[0017] In this embodiment, the DC current of the electrolytic cell is obtained by a Hall-type DC current transmitter installed on the DC bus side of the electrolytic cell, the mass flow rate of the circulating electrolyte is obtained by a Coriolis mass flow meter installed on the outlet pipeline of the electrolyte heat exchanger, the opening degree of the heat exchange regulating valve is obtained by a valve position transmitter configured on the heat exchange regulating valve, and the pipeline-side reference temperature is obtained by a resistance heating element attached to the pipeline side of the outlet temperature sensor mounting interface.

[0018] The resistance heating element uses a thin-film metal heating element with resistance temperature characteristics. A 0.05mm thick thermally conductive insulation layer is set between the resistance heating element and the outer wall of the circulation pipeline. The outside of the resistance heating element is covered with a 20mm thick removable insulation layer. The resistance heating element uses a 4-wire connection, with 2 wires connected to the constant power drive circuit and the other 2 wires connected to the voltage measurement circuit to reduce the influence of the power supply wire resistance on the resistance measurement results.

[0019] In a preferred embodiment, the resistance-temperature characteristics of the resistance heating element are obtained on a calibration pipe section with the same material, wall thickness, thermally conductive insulation layer, and adhesion pressure as the actual circulation pipeline. A calibrated thin-film platinum resistance thermometer is installed at the contact position between the resistance heating element and the pipe wall in the calibration pipe section. The calibration pipe section is sequentially adjusted to 40°C, 50°C, 60°C, 70°C, 80°C, 90°C, and 100°C. At each calibration temperature, after the temperature change of the thin-film platinum resistance thermometer does not exceed 0.02°C for 60 seconds, the voltage across the resistance heating element and the current through the resistance heating element are collected. The resistance value is obtained according to the ratio of voltage to current, and the resistance value is correlated with the pipe wall temperature measured by the thin-film platinum resistance thermometer to form a piecewise linear correspondence between the resistance value and the reference temperature on the pipeline side.

[0020] The above calibration is performed with the resistance heating element in the attached state. Therefore, the obtained resistance temperature characteristics include the influence of the thermally conductive insulation layer, the attachment pressure, and the heat dissipation conditions of the pipe wall on the average temperature of the resistance heating element. This avoids the temperature deviation caused by directly replacing the pipe-side reference temperature with the calibration results of the suspended heating element. In the actual local thermal excitation process, the control station performs a linear conversion between two adjacent calibrated resistance values ​​based on the current resistance value to obtain the pipe-side reference temperature at the corresponding sampling time.

[0021] Before applying local thermal excitation, the control station selects a 60-second continuous operating condition judgment section and obtains the maximum and minimum values ​​of the electrolytic cell DC current, circulating electrolyte mass flow rate, heat exchange regulating valve opening, and pipeline-side reference temperature within the operating condition judgment section. When the difference between the maximum and minimum values ​​of the electrolytic cell DC current is not greater than 0.5% of the average value of the electrolytic cell DC current within the operating condition judgment section, the difference between the maximum and minimum values ​​of the circulating electrolyte mass flow rate is not greater than 0.5% of the average value of the circulating electrolyte mass flow rate within the operating condition judgment section, the difference between the maximum and minimum values ​​of the heat exchange regulating valve opening is not greater than 0.3 percentage points, and the difference between the maximum and minimum values ​​of the pipeline-side reference temperature is not greater than 0.05℃, the operating condition judgment section is determined to meet the local thermal excitation conditions.

[0022] The temperature range between the minimum and maximum reference temperatures on the pipeline side within a continuous 60 seconds that meet the local thermal excitation conditions is defined as the stable range. To exclude continuous heating or cooling conditions, within a continuous 60 seconds, the average reference temperature on the pipeline side for the first 10 seconds is compared with the average reference temperature for the last 10 seconds. The pipeline side reference temperature can be considered to be in the stable range only when the absolute value of the difference between the two is not greater than 0.03℃.

[0023] In this embodiment, the average DC current of the electrolytic cell over 60 seconds is 6000A, with a variation range of 5992A to 6008A; the average mass flow rate of the circulating electrolyte is 33.0kg / s, with a variation range of 32.94kg / s to 33.07kg / s; the average opening degree of the heat exchange regulating valve is 40.0%, with a variation range of 39.9% to 40.1%; and the reference temperature on the pipeline side varies from 78.01℃ to 78.05℃. The difference between the average reference temperature on the pipeline side in the first 10 seconds and the average reference temperature on the pipeline side in the last 10 seconds is 0.01℃. Therefore, 78.01℃ to 78.05℃ is determined as the stable range.

[0024] Under the above operating conditions, the control station maintains the DC current control command of the electrolytic cell, the speed control command of the circulating pump, and the opening command of the heat exchange regulating valve unchanged, and applies a constant power pulse of 48W for 30s to the resistance heating element; the constant power drive circuit adjusts the output voltage according to the resistance heating element voltage and current obtained in each 100ms acquisition cycle, so that the product of voltage and current is between 47.8W and 48.2W in each acquisition cycle; the power amplitude, duration, installation position of the resistance heating element, thickness of the thermally conductive insulation layer, thickness of the heat insulation layer, and sampling interval of the constant power pulse are the same as those used when establishing the reference contact sample.

[0025] The control station continuously acquires the voltage and current of the resistance heating element within the range from 3 seconds before the start of local thermal excitation to 60 seconds after the stop of local thermal excitation. The resistance value of the resistance heating element is obtained based on the average voltage and average current within each 1-second sampling segment. Then, the pipeline-side reference temperature is obtained according to the piecewise linear correspondence, and the pipeline-side reference temperatures are arranged according to the sampling time to form a pipeline-side reference temperature response.

[0026] Preferably, the control station simultaneously acquires the sensor temperature output by the outlet temperature sensor, arranges the sensor temperatures according to the same sampling time, and forms a sensor temperature response; the pipeline-side reference temperature response reflects the temperature change generated by local thermal excitation on the pipeline side at the outlet temperature sensor mounting interface, and the sensor temperature response reflects the temperature change after the temperature change is transmitted to the outlet temperature sensor through the mounting contact interface.

[0027] S1.2. The minimum and maximum values ​​of the pipeline-side reference temperature and the minimum and maximum values ​​of the sensor temperature at three consecutive sampling times before excitation are used to construct the reference temperature fluctuation range and the sensor temperature fluctuation range, respectively. The first sampling time when the pipeline-side reference temperature is higher than the upper limit of the reference temperature fluctuation range for two consecutive sampling times is determined as the reference response start point, and the first sampling time when the sensor temperature is higher than the upper limit of the sensor temperature fluctuation range for two consecutive sampling times is determined as the sensor response start point.

[0028] In a preferred embodiment, the start time of local thermal excitation is determined to be 0s, the first three sampling times before local thermal excitation are determined to be -3s, -2s and -1s respectively, and the sampling times after local thermal excitation are determined to be 0s, 1s, 2s and 3s respectively; the pipeline-side reference temperature response and the sensor temperature response use the same sampling sequence number, and the same sampling sequence number corresponds to the same physical time.

[0029] In this embodiment, the pipeline-side reference temperatures at -3s, -2s, and -1s are 78.01℃, 78.03℃, and 78.02℃, respectively, resulting in a reference temperature fluctuation range of 78.01℃ to 78.03℃; the sensor temperatures at the same three sampling times are 77.95℃, 77.97℃, and 77.96℃, respectively, resulting in a sensor temperature fluctuation range of 77.95℃ to 77.97℃.

[0030] After local thermal excitation, the reference temperature on the pipeline side is continuously higher than 78.03℃ for 2s and 3s, so 2s is determined as the reference response start point; the sensor temperature is continuously higher than 77.97℃ for 7s and 8s, so 7s is determined as the sensor response start point; when only one sampling time is higher than the upper limit of the corresponding fluctuation range, it is not determined as the response start point, so as to eliminate the erroneous judgment caused by a single temperature measurement fluctuation.

[0031] S1.3 Calculate the response delay between the reference response start point and the sensor response start point; using the average reference temperature on the pipeline side before excitation and the average sensor temperature as benchmarks, calculate the reference temperature rise peak and the sensor temperature rise peak, and determine the normalized amplitude attenuation by subtracting the ratio of the sensor temperature rise peak to the reference temperature rise peak.

[0032] Specifically, the method for calculating response delay includes: subtracting the sampling number of the reference response start point from the sampling number of the sensor response start point to obtain the sampling number difference, and determining the response delay by multiplying the sampling number difference by the sampling interval.

[0033] In this embodiment, the sensor response start point is 7s, the reference response start point is 2s, the difference between their sampling sequence numbers is 5, and the response delay is 5s.

[0034] In a preferred embodiment, the control station calculates the average value of the pipeline-side reference temperature and the average value of the sensor temperature at the three sampling times before the local thermal excitation; within the evaluation period from the start of the local thermal excitation to 60 seconds after the local thermal excitation stops, the maximum value of the pipeline-side reference temperature and the maximum value of the sensor temperature are selected respectively; the reference temperature rise peak is obtained by subtracting the average value of the pipeline-side reference temperature before excitation from the maximum value of the pipeline-side reference temperature; the sensor temperature rise peak is obtained by subtracting the average value of the sensor temperature before excitation from the maximum value of the sensor temperature.

[0035] In this embodiment, the average reference temperature on the pipeline side before excitation is 78.02℃, and the maximum reference temperature on the pipeline side in the evaluation section is 80.62℃, resulting in a reference temperature rise peak of 2.60℃; the average sensor temperature before excitation is 77.96℃, and the maximum sensor temperature in the evaluation section is 79.63℃, resulting in a sensor temperature rise peak of 1.67℃; dividing the sensor temperature rise peak by the reference temperature rise peak yields 0.642, and subtracting 0.642 from 1 yields a normalized amplitude attenuation of 0.358.

[0036] When the reference temperature rise peak is below 0.50℃, the temperature rise amplitude generated by local thermal excitation is insufficient to distinguish it from temperature measurement fluctuations. The result of this local thermal excitation is discarded, and a constant power pulse is applied again after the reference temperature on the pipeline side re-enters the stable range.

[0037] S1.4 The contact response characteristics consist of response delay and normalized amplitude decay.

[0038] In a preferred embodiment, the control station arranges the response delay and normalized amplitude decay in a fixed order, with the response delay preceding the normalized amplitude decay, to form a contact response characteristic; the response delay is retained up to 0.1s, and the normalized amplitude decay is retained up to 0.001s.

[0039] In this embodiment, the contact response characteristics are a response delay of 5.0s and a normalized amplitude decay of 0.358; the pipeline-side reference temperature of 78.02℃, the circulating electrolyte mass flow rate of 33.0kg / s, and the pulse power of 48W before local thermal excitation are stored as test conditions for the contact response characteristics.

[0040] S2. Compare the contact response characteristics with the baseline contact response characteristics at the same temperature and flow rate to determine the installation contact state and select the installation temperature measurement heat transfer model. Note that the following points should be noted in this step: S2.1. Using the reference temperature on the pipeline side and the circulating electrolyte flow rate before local thermal excitation as the matching temperature and matching flow rate, respectively, select the reference contact response features with the same temperature and flow rate as the matching temperature and matching flow rate from the closely contacted samples. If there are no reference contact response features, select four sets of adjacent reference contact response features with temperature and flow rate around the matching temperature and matching flow rate, and obtain the reference contact response features with the same temperature and flow rate through bilinear interpolation.

[0041] In a preferred embodiment, the tight contact sample is obtained using the same model of temperature sensor, the same structure of temperature measuring sleeve, the same thermally conductive medium, the same resistive heating element, the same thermally conductive insulation layer, and the same insulation structure as the outlet temperature sensor to be tested. When forming the tight contact sample, the tightening torque of the outlet temperature sensor mounting thread is adjusted to 6 N·m, the thickness of the thermally conductive medium is controlled between 0.06 mm and 0.10 mm, and the contact area of ​​the mounting interface is checked using a pressure-sensitive contact film. When the effective contact area of ​​the mounting interface accounts for 95% of the total area of ​​the mounting interface, and the offset between the axis of the outlet temperature sensor and the axis of the temperature measuring sleeve is not greater than 0.2 mm, the mounting state is determined to be a tight contact state.

[0042] Close-contact samples were obtained under combined operating conditions of pipeline-side reference temperatures of 70℃, 75℃, 80℃, and 85℃, and circulating electrolyte mass flow rates of 30kg / s, 35kg / s, and 40kg / s. For each combined operating condition, a local thermal excitation of 48W for 30s was applied five times, with an interval of no less than 300s between adjacent local thermal excitations. The next local thermal excitation was initiated after the pipeline-side reference temperature returned to the corresponding stable range.

[0043] The control station determines the average value of the pipeline-side reference temperature at the three sampling times prior to the current local thermal excitation as the matching temperature, and the average value of the circulating electrolyte mass flow rate at the same three sampling times as the matching flow rate. When there are samples in close contact where the absolute value of the temperature difference between the sample and the matching temperature is no greater than 0.01℃ and the absolute value of the flow rate difference between the sample and the matching flow rate is no greater than 0.05kg / s, the median response delay and the median normalized amplitude decay of the corresponding samples are used to form the benchmark contact response characteristics at the same temperature and flow rate.

[0044] If there are no closely contacted samples that meet the above conditions, select the temperature that is lower than the matching temperature and closest to the matching temperature, the temperature that is higher than the matching temperature and closest to the matching temperature, the flow rate that is lower than the matching flow rate and closest to the matching flow rate, and the flow rate that is higher than the matching flow rate and closest to the matching flow rate. These four sets of adjacent operating conditions are formed by two adjacent temperatures and two adjacent flow rates.

[0045] Bilinear interpolation is performed for response delay and normalized amplitude attenuation. Specifically, at a lower temperature, the first intermediate feature value is obtained based on the position of the matched flow rate between the lower and higher flow rates. Then, at a higher temperature, the same process is used to obtain the second intermediate feature value. Subsequently, based on the position of the matched temperature between the lower and higher temperatures, the first and second intermediate feature values ​​are interpolated to obtain feature values ​​corresponding to the matched temperature and matched flow rate. After the above processing is performed on response delay and normalized amplitude attenuation, the reference contact response features are formed.

[0046] In this embodiment, the matched temperature is 78.02℃, the matched flow rate is 33.0kg / s, and 75℃, 80℃, 30kg / s, and 35kg / s are selected to form four adjacent operating conditions. The response delay and normalized amplitude decay of the 75℃ and 30kg / s operating conditions are 2.8s and 0.205, respectively; the response delay and normalized amplitude decay of the 75℃ and 35kg / s operating conditions are 2.9s and 0.212, respectively; the response delay and normalized amplitude decay of the 80℃ and 30kg / s operating conditions are 3.1s and 0.224, respectively; and the response delay and normalized amplitude decay of the 80℃ and 35kg / s operating conditions are 3.2s and 0.231, respectively. After bilinear interpolation, the baseline response delay is 3.04s and the baseline normalized amplitude decay is 0.221.

[0047] S2.2. Subtract the response delay in the reference contact response characteristics from the response delay in the contact response characteristics to obtain the response delay deviation; subtract the normalized amplitude attenuation in the reference contact response characteristics from the normalized amplitude attenuation in the contact response characteristics to obtain the amplitude attenuation deviation.

[0048] S2.3 Calculate the median of response delay deviation and the median of amplitude attenuation deviation for the closely contacted sample, the sample with increased contact thermal resistance, and the sample with partial separation, respectively; determine the midpoint of the median of response delay deviation for the closely contacted sample and the sample with increased contact thermal resistance as the first boundary of response delay, and determine the midpoint of the median of response delay deviation for the sample with increased contact thermal resistance and the sample with partial separation as the second boundary of response delay; determine the first boundary and the second boundary of amplitude attenuation in the same way.

[0049] In a preferred embodiment, the samples with increased contact thermal resistance and the samples with partial detachment are obtained using the same sensor structure, local thermal excitation conditions, and operating conditions as the samples with close contact. When forming the sample with increased contact thermal resistance, a mica heat insulation sheet with a thickness of 0.20 mm and a thermal conductivity of 0.30 W / (m·K) is set in the complete installation contact interface, and the tightening torque of the installation thread is adjusted to 6 N·m to keep the installation interface in a complete and compacted state, thereby increasing the installation contact thermal resistance.

[0050] When forming a partially detached sample, an arc-shaped spacer with a thickness of 0.30 mm is set on one side of the installation interface so that 30% of the area of ​​the installation interface forms an air gap, while the remaining 70% of the area remains in contact. When the area of ​​the air gap is confirmed to be between 28% and 32% of the total area of ​​the installation interface by the pressure-sensitive contact film, the installation state is determined to be a partially detached state.

[0051] Furthermore, samples of close contact, samples of increased contact thermal resistance, and samples of partial detachment were obtained separately with the same combination of temperature and flow rate. For each set of contact response characteristics, the close contact reference contact response characteristics with the same temperature and flow rate were selected, and the response delay deviation and amplitude attenuation deviation were obtained respectively.

[0052] All response delay deviations under the same installation condition are arranged in ascending order of value; when the number of samples is odd, the value in the middle position of the arrangement is selected as the median of the response delay deviation; when the number of samples is even, the arithmetic mean of the two middle values ​​is selected as the median of the response delay deviation; the median of the amplitude attenuation deviation is processed in the same way.

[0053] In this embodiment, the median response delay deviations for the closely contacted sample, the sample with increased contact thermal resistance, and the sample with partial detachment are 0.12s, 2.30s, and 6.70s, respectively. The arithmetic mean of 0.12s and 2.30s, 1.21s, is determined as the first boundary of the response delay, and the arithmetic mean of 2.30s and 6.70s, 4.50s, is determined as the second boundary of the response delay.

[0054] The median amplitude attenuation deviations for the close contact sample, the sample with increased contact thermal resistance, and the sample with partial detachment were 0.010, 0.146, and 0.352, respectively. The arithmetic mean of 0.010 and 0.146, 0.078, was determined as the first boundary of amplitude attenuation, and the arithmetic mean of 0.146 and 0.352, 0.249, was determined as the second boundary of amplitude attenuation.

[0055] S2.4 Determine the contact state level to which the response delay deviation and amplitude attenuation deviation belong respectively, and determine the higher of the two contact state levels as the installation contact state in the order of tight contact level, increased contact thermal resistance level and partial separation level from low to high.

[0056] Specifically, the determination of the contact status level includes: When the response delay deviation is below the first boundary of the response delay, it is determined to be a close contact level; when the response delay deviation reaches the first boundary of the response delay but is below the second boundary of the response delay, it is determined to be a level of increased contact thermal resistance; when the response delay deviation reaches the second boundary of the response delay, it is determined to be a level of partial separation. When the amplitude attenuation deviation is below the first boundary of amplitude attenuation, it is determined to be a tight contact level; when the amplitude attenuation deviation reaches the first boundary of amplitude attenuation but is below the second boundary of amplitude attenuation, it is determined to be a level of increased contact thermal resistance; when the amplitude attenuation deviation reaches the second boundary of amplitude attenuation, it is determined to be a level of partial separation.

[0057] S2.5 Substitute the pipeline-side reference temperature response, sensor temperature response, and ambient temperature of the sample belonging to the installation contact state into the heat balance equation of the sensor thermal node. Under the constraint that the installation contact thermal resistance, sensor heat dissipation thermal resistance, and sensor heat capacity are all greater than zero, perform nonlinear least squares identification with the condition that the sum of the squared deviations between the calculated sensor temperature value and the measured sensor temperature value is minimized.

[0058] In a preferred embodiment, samples with the same structure as the current outlet temperature sensor and whose sample temperature and flow ranges cover the current operating conditions are selected from the samples with increased contact thermal resistance. Each sample group includes a pipeline-side reference temperature response, a sensor temperature response, and an ambient temperature. The ambient temperature is obtained by a shielded platinum resistance temperature sensor located 100 mm outside the outlet temperature sensor. A heat insulation baffle is provided between the ambient temperature sensor and the resistance heating element. The ambient temperature is aligned with the pipeline-side reference temperature response and the sensor temperature response at a 1-second sampling interval.

[0059] For example, the heat balance relationship of the sensor hot node is expressed as:

[0060] in, For sensor heat capacity, Let be the sensor temperature at time t. Let be the pipeline-side reference temperature at time t. Let be the ambient temperature at time t. To install contact thermal resistors, The thermal resistance of the sensor is used for heat dissipation, and t is the continuous time. The sensor temperature change rate is represented by s, where s represents the sensor thermal node; p represents the pipeline-side thermal node; a represents the ambient temperature boundary; and c represents the installation contact heat transfer path.

[0061] The control station uses the measured sensor temperature at the first sampling moment of each sample group as the initial temperature, and performs numerical recursion on the heat balance relationship of the sensor thermal nodes according to a calculation step of 1 second, and obtains the calculated sensor temperature value at each sampling moment in sequence.

[0062] For example, the identification conditions for mounting contact thermal resistance, sensor heat dissipation thermal resistance, and sensor thermal capacity are expressed as follows:

[0063] Where k is the sampling sequence number; N is the number of sampling points participating in the current sample identification; This represents the measured sensor temperature at the k-th sampling time. Let m be the calculated sensor temperature at the k-th sampling time, and m be the measured value.

[0064] In a specific processing step, the search range for the installation contact thermal resistance is set to 0.05 K / W to 5.00 K / W, the search range for the sensor heat dissipation thermal resistance is set to 5.0 K / W to 50.0 K / W, and the search range for the sensor heat capacity is set to 1.0 J / K to 30.0 J / K. The three thermal parameters are updated using a damped Gauss-Newton method with boundary constraints. When the sum of the squared deviations and the change ratios of two adjacent iterations are less than 0.001, or when the number of iterations reaches 200, the current sample identification ends.

[0065] When any identification result reaches the upper or lower limit of the corresponding search range, the current sample is identified as a boundary identification sample and is not included in the median calculation in step S2.6; when the root mean square deviation between the calculated sensor temperature value and the measured sensor temperature value exceeds 0.10℃, the current sample is identified as a model-unsuitable sample and is not included in the median calculation in step S2.6.

[0066] In this embodiment, five groups of samples with increased contact thermal resistance were selected for identification. The obtained installation contact thermal resistances were 1.28 K / W, 1.34 K / W, 1.31 K / W, 1.36 K / W, and 1.33 K / W, respectively; the sensor heat dissipation thermal resistances were 17.4 K / W, 17.8 K / W, 17.6 K / W, 18.1 K / W, and 17.9 K / W, respectively; and the sensor heat capacities were 11.2 J / K, 11.6 J / K, 11.4 J / K, 11.7 J / K, and 11.5 J / K, respectively. The root mean square deviation between the calculated and measured sensor temperatures for each group of samples was 0.032℃ to 0.047℃, and none of the three thermal parameters reached the corresponding search range boundary.

[0067] S2.6. Take the median of the identified installation contact thermal resistance, sensor heat dissipation thermal resistance, and sensor thermal capacity respectively, and load the installation contact thermal resistance, sensor heat dissipation thermal resistance, and sensor thermal capacity into the two-node thermal resistance-thermal capacity model to obtain the installation temperature measurement heat transfer model.

[0068] The two-node thermal resistance-thermal capacity model includes a pipeline-side thermal node and a sensor thermal node. The pipeline-side thermal node and the sensor thermal node are connected by an installation contact thermal resistance. The sensor thermal node is associated with the sensor thermal capacity and is connected to the ambient temperature boundary through the sensor heat dissipation thermal resistance. The installation temperature measurement heat transfer model takes the pipeline-side temperature and ambient temperature as inputs and the sensor temperature as output.

[0069] In a preferred embodiment, the identified mounting contact thermal resistance, sensor heat dissipation thermal resistance, and sensor thermal capacity are arranged in ascending order of value, and the median of each group of parameters is taken to obtain mounting contact thermal resistance of 1.33 K / W, sensor heat dissipation thermal resistance of 17.8 K / W, and sensor thermal capacity of 11.5 J / K.

[0070] The installation contact thermal resistance, sensor heat dissipation thermal resistance, and sensor thermal capacity are loaded into a two-node thermal resistance-thermal capacity model consisting of a pipeline-side thermal node, a sensor thermal node, an installation contact heat transfer path, and an ambient heat dissipation path. The pipeline-side thermal node receives the pipeline-side temperature, the sensor thermal node corresponds to the thermal state of the outlet temperature sensor, the pipeline-side thermal node and the sensor thermal node are connected through the installation contact thermal resistance, the sensor thermal node is connected to the ambient temperature boundary through the sensor heat dissipation thermal resistance, and the sensor thermal capacity is associated with the sensor thermal node.

[0071] For example, after discrete recursion according to a 1-second sampling interval, the heat transfer model for temperature measurement is represented as follows:

[0072] in, For the first The sensor temperature calculation value at each sampling time. For the first Sensor temperature at each sampling time, For the first Pipeline side temperature at each sampling time, For the first The ambient temperature at each sampling time The sampling interval is 1 second in this embodiment; For the sensor's heat capacity, this embodiment uses 11.5 J / K; For the installation contact thermal resistance, this embodiment uses 1.33 K / W; The thermal resistance for heat dissipation of the sensor is set to 17.8 K / W in this embodiment.

[0073] When performing the initial calculation of the temperature measurement and heat transfer model, the measured temperature value of the outlet temperature sensor at the starting point of the calculation is used as the initial value of the sensor temperature. When using a 1-second sampling interval, the sampling interval is small enough compared with the sensor's heat capacity and the minimum thermal time scale formed by the two heat transfer paths, so that the discrete recursive process will not produce numerical divergence.

[0074] S3. The sensor temperature response under varying net heat input is retrieved using an installed temperature measurement heat transfer model to obtain the pipeline-side temperature response. Based on the net heat input and pipeline-side temperature response, the electrolytic cell thermal response model is identified, and the thermal inertia characteristics of the cell-electrolyte relationship are extracted. Note that the following points should be noted in this step: S3.1. The product of the difference between the total voltage of the electrolytic cell and the thermal neutral voltage of the electrolytic cell and the DC current of the electrolytic cell is determined as the electrochemical heat generation; the product of the circulating electrolyte mass flow rate, the specific heat capacity of the electrolyte, and the difference between the electrolyte inlet temperature and the electrolyte outlet temperature of the electrolyte heat exchanger is determined as the heat removal amount; the net heat input is obtained by subtracting the heat removal amount from the electrochemical heat generation.

[0075] In this embodiment, the total thermal neutral voltage of the electrolyzer refers to the sum of the thermal neutral voltages of all series electrolytic cells corresponding to the enthalpy of the water decomposition reaction under the current electrolyte temperature and electrolyzer operating pressure. When the total voltage of the electrolyzer is equal to the total thermal neutral voltage of the electrolyzer, the electrical energy input to the electrolyzer is balanced with the energy absorbed by the water decomposition reaction, without considering the external heat dissipation of the electrolyzer.

[0076] The total thermal neutral voltage of the electrolytic cell is obtained based on the number of effective series-connected cells and the thermal neutral voltage of a single electrolytic cell; the number of effective series-connected cells is determined according to the electrolytic cell's factory technical documents and DC wiring structure; in this embodiment, the number of effective series-connected cells is 360.

[0077] The control station uses the arithmetic mean of the electrolyte temperature at the inlet and outlet of the electrolyzer as the average temperature of the electrolyzer, and the pressure measurement value of the gas-liquid separator as the operating pressure of the electrolyzer. It also determines the thermal neutral voltage of a single electrolysis cell based on the thermodynamic data table of the enthalpy of water decomposition reaction and temperature and pressure. When the average temperature or operating pressure of the electrolyzer is between two adjacent data points, a linear conversion is performed between the adjacent data points.

[0078] In this embodiment, the average temperature of the electrolytic cell is 80.0℃, the operating pressure of the electrolytic cell is 0.30MPa, the corresponding thermal neutral voltage of a single electrolytic cell is 1.480V, and the total thermal neutral voltage of the electrolytic cell corresponding to 360 series-connected electrolytic cells is 532.8V; the current total voltage of the electrolytic cell is 756.0V, the DC current of the electrolytic cell is 6000A, and the difference between the total voltage of the electrolytic cell and the total thermal neutral voltage of the electrolytic cell is 223.2V. Multiplying 223.2V by 6000A, we obtain an electrochemical heat generation of 1339.2kW.

[0079] The circulating electrolyte mass flow rate is 33.0 kg / s. The specific heat capacity of the electrolyte corresponding to the current electrolyte concentration and temperature is 3.75 kJ / (kg·K). The electrolyte inlet temperature of the electrolyte heat exchanger is 82.5℃, and the electrolyte outlet temperature of the electrolyte heat exchanger is 73.4℃, with a temperature difference of 9.1 K. Multiplying the circulating electrolyte mass flow rate, the specific heat capacity of the electrolyte, and the temperature difference, we obtain a heat transfer output of 1126.1 kW. Subtracting the heat transfer output of 1126.1 kW from the electrochemical heat generation of 1339.2 kW, we obtain a net heat input of 213.1 kW.

[0080] S3.2 When the resistive heating element is not powered on, align the sensor temperature response, ambient temperature and net heat input formed by the temperature sensor during the period of net heat input change according to the sampling time, and divide the difference of sensor temperature at adjacent sampling times by the sampling interval to obtain the sensor temperature change rate.

[0081] In a preferred embodiment, the electrolytic cell thermal response is identified during a continuous operating period in which the resistance heating element is de-energized for at least 120 seconds, in order to reduce the impact of residual heat from local thermal excitation on the pipeline temperature response. During the continuous operating period, the change in net heat input relative to the stable value before the change is not less than 80 kW, and the change lasts for at least 300 seconds.

[0082] The control station establishes a unified sampling time sequence with 1-second intervals. Within each 1-second sampling segment, the arithmetic mean of the DC current and total voltage of the electrolytic cell obtained in a 100ms acquisition cycle is calculated, and the electrochemical heat generation at the corresponding sampling time is obtained based on the arithmetic mean. The circulating electrolyte mass flow rate, electrolyte inlet temperature of the electrolyte heat exchanger, electrolyte outlet temperature of the electrolyte heat exchanger, sensor temperature, and ambient temperature are measured using the values ​​with the smallest time difference from the unified sampling time, with a time difference not exceeding 0.2 seconds.

[0083] When two adjacent temperature measurement times are located on either side of a unified sampling time, a linear conversion is performed based on the time distance between the unified sampling time and the two adjacent temperature measurement times; when a single measurement point is missing, it is linearly supplemented using adjacent valid measurement points; when two or more consecutive measurement points are missing, the identification segment containing the missing location is discarded.

[0084] The sampling interval is 1 second. The control station subtracts the sensor temperature from the previous sampling time from the current sampling time, and then divides by 1 second to obtain the sensor temperature change rate at the current sampling time; for example, if the sensor temperature at the 101st second is 79.24℃ and the sensor temperature at the 102nd second is 79.27℃, then the sensor temperature change rate corresponding to the 102nd second is 0.03K / s.

[0085] S3.3. The total heat flux of the sensor's thermal node is obtained by multiplying the sensor's heat capacity by the sensor's temperature change rate, adding the difference between the sensor's temperature and the ambient temperature by the ratio obtained by dividing the sensor's heat dissipation thermal resistance; the total heat flux of the sensor's thermal node is then added to the sensor's temperature by multiplying the total heat flux of the sensor's thermal node by the installation contact thermal resistance, and the pipe-side temperature is obtained by arranging the pipe-side temperatures according to the sampling time.

[0086] The control station uses the sensor heat capacity of 11.5 J / K, sensor heat dissipation thermal resistance of 17.8 K / W, and installation contact thermal resistance of 1.33 K / W obtained in step S2 to perform pipeline-side temperature inversion.

[0087] For each sampling moment, first multiply the sensor's heat capacity by the sensor's temperature change rate to obtain the change in heat storage at the sensor's thermal node; then divide the difference between the sensor's temperature and the ambient temperature by the sensor's heat dissipation thermal resistance to obtain the heat flow from the sensor's thermal node to the environment; add the change in heat storage to the heat flow to obtain the total heat flow from the pipeline side to the sensor's thermal node; multiply the total heat flow by the installation contact thermal resistance and add it to the current sensor temperature to obtain the pipeline side temperature at the current sampling moment.

[0088] For example, the current sensor temperature is 79.27℃, the ambient temperature is 29.00℃, the sensor temperature change rate is 0.03K / s, the sensor heat capacity is 11.5J / K, the sensor heat dissipation thermal resistance is 17.8K / W, and the mounting contact thermal resistance is 1.33K / W; the heat storage change of the sensor hot node is 0.345W, the heat flow transferred from the sensor hot node to the environment is 2.824W, and the total heat flow transferred from the pipeline side to the sensor hot node is 3.169W, corresponding to a pipeline side temperature of 83.49℃.

[0089] The control station repeats the above processing for each sampling time within the identification section and forms a pipeline side temperature response based on the obtained pipeline side temperatures arranged according to the sampling times. To eliminate non-physical results in the inversion process, when the pipeline side temperature at any sampling time is more than 5°C lower than the electrolyte temperature at the outlet of the electrolyte heat exchanger or more than 5°C higher than the upper limit of the electrolytic cell operating temperature, the sensor temperature change rate, ambient temperature, and sensor status at the corresponding sampling time are checked. If a measurement abnormality is confirmed, the identification section containing that sampling time is discarded.

[0090] S3.4. Using the net heat input and pipe-side temperature at the identification starting point as benchmarks, respectively, generate the net heat input increment and pipe-side temperature rise. Set candidate pure lag durations with different integer multiple sampling intervals. For each candidate pure lag duration, take the pipe-side temperature rise at the current sampling time as the dependent variable, the pipe-side temperature rise corresponding to the previous sampling time adjacent to the current sampling time, and the net heat input increment at the sampling time corresponding to the candidate pure lag duration traced back from the current sampling time as independent variables. Perform least squares fitting under the condition that the discrete attenuation coefficient is greater than zero and less than one, and the heat input response coefficient is non-negative, to obtain the discrete attenuation coefficient, heat input response coefficient, and fitting residual.

[0091] Specifically, the first sampling time when the net heat input deviates from the average value of the previous 30 seconds for three consecutive sampling times and the direction of deviation is consistent is determined as the identification starting point; the net heat input at the identification starting point is used as the net heat input benchmark, and the pipeline side temperature at the identification starting point is used as the pipeline side temperature benchmark.

[0092] For each sampling moment after the identification starting point, the net heat input increment is formed by subtracting the net heat input benchmark from the current net heat input; the pipe side temperature rise is formed by subtracting the pipe side temperature benchmark from the current pipe side temperature; both the net heat input increment and the pipe side temperature rise are based on the identification starting point as zero benchmark, so a constant bias term is no longer introduced separately in the discrete identification relationship.

[0093] The upper limit of the candidate pure lag time is determined based on the pipeline volume between the electrolyzer outlet and the installation position of the outlet temperature sensor and the medium residence time corresponding to the lowest circulating electrolyte mass flow rate. In this embodiment, the medium residence time does not exceed 27s, the medium residence time is rounded up to 30s, and 0s, 1s, 2s up to 30s are set as candidate pure lag times.

[0094] For each candidate pure time delay, the fitting data is organized row by row according to the sampling time. Each row of fitting data uses the pipe-side temperature rise at the current sampling time as the dependent variable, the pipe-side temperature rise corresponding to the previous sampling time adjacent to the current sampling time as the first independent variable, and the net heat input increment corresponding to the sampling time after tracing back the candidate pure time delay from the current sampling time as the second independent variable.

[0095] For example, when the candidate pure lag duration is 12s, the pipe-side temperature rise at 200s is paired with the pipe-side temperature rise at 199s, and the net heat input increment at 188s is selected to participate in the fitting; when the front sampling point corresponding to the candidate pure lag duration lacks the net heat input increment traced back, the front sampling point does not participate in the fitting of the current candidate pure lag duration.

[0096] For each candidate pure time delay, the control station performs constrained least squares fitting, with the discrete attenuation coefficient limited to a range greater than 0 and less than 1, and the thermal input response coefficient limited to a range not less than 0. Initial coefficients are first obtained using unconstrained least squares. If the initial coefficients do not meet the range, bounded least squares are used to obtain the discrete attenuation coefficient and thermal input response coefficient again.

[0097] Subsequently, the fitted values ​​of the pipeline side temperature rise are calculated one by one according to the discrete attenuation coefficient and the thermal input response coefficient. The difference between the measured value of the pipeline side temperature rise and the fitted value of the pipeline side temperature rise is squared and accumulated to obtain the fitting residual of the corresponding candidate pure lag time.

[0098] In this embodiment, when the candidate pure time delay is 11s, 12s, and 13s, the fitting residuals are 1.04℃, respectively. 2 0.92℃ 2 and 0.99℃ 2 The fitting residuals of the remaining candidate pure time delays are all greater than 0.99℃. 2 When the candidate pure time delay is 12s, the discrete attenuation coefficient is 0.99762, the thermal input response coefficient is 0.00001142℃ / kW, and the fitting residual is 0.92℃. 2 Therefore, 12s was determined as the pure time delay.

[0099] S3.5. The candidate pure lag time with the smallest fitting residual is determined as the pure lag time. The sampling interval is divided by the absolute value of the natural logarithm of the discrete attenuation coefficient to obtain the electrolytic cell time constant. The thermal input response coefficient is divided by the difference obtained by subtracting the discrete attenuation coefficient to obtain the steady-state thermal gain. The electrolytic cell thermal response model is composed of the electrolytic cell time constant, steady-state thermal gain and pure lag time.

[0100] For example, the electrolytic cell time constant is calculated based on the discrete attenuation coefficient and the sampling interval, and the calculation relationship is expressed as follows:

[0101] in, The time constant of the electrolytic cell. The sampling interval is 1 second in this embodiment; The discrete attenuation coefficient is taken as 0.99762 in this embodiment; Represents the natural logarithm operation.

[0102] The electrolytic cell time constant calculated based on the above values ​​is 419.7 s.

[0103] For example, the steady-state thermal gain is calculated based on the thermal input response coefficient and the discrete attenuation coefficient, and the calculation relationship is expressed as follows:

[0104] in, For steady-state thermal gain, The thermal input response coefficient is taken as 0.00001142℃ / kW in this embodiment.

[0105] The steady-state thermal gain calculated based on the above values ​​is 0.00480℃ / kW.

[0106] In this embodiment, the electrolytic cell time constant of 419.7s is obtained based on the sampling interval of 1s and the discrete attenuation coefficient of 0.99762; the steady-state thermal gain of 0.00480℃ / kW is obtained based on the thermal input response coefficient of 0.00001142℃ / kW and the discrete attenuation coefficient of 0.99762.

[0107] For example, the thermal response model of a first-order inertial pure time-delay electrolyzer, consisting of the electrolyzer time constant, steady-state thermal gain, and pure time delay, is expressed as follows:

[0108] in, The relationship between the incremental net heat input and the temperature rise on the pipeline side is represented by L, which is the pure time delay, taken as 12s in this embodiment; s is the Laplace transform variable; e is the natural constant; the negative exponent indicates that the incremental net heat input enters the first-order inertial element after the pure time delay.

[0109] Furthermore, the electrolytic cell thermal response model uses the net heat input increment as input and the pipeline side temperature rise as output. During the simulation calculation, the pipeline side temperature rise is added to the pipeline side temperature at the simulation starting point to obtain the pipeline side temperature. After the net heat input increment changes, the pipeline side temperature rise caused by this net heat input change does not enter the model output within 12s. After 12s, it gradually approaches the temperature rise value determined by the steady-state thermal gain according to the electrolytic cell time constant of 419.7s.

[0110] S3.6. The ratio of the electrolytic cell time constant to the steady-state thermal gain is determined as the equivalent heat capacity of the cell-electrolyte. The thermal inertia characteristic of the cell-electrolyte is composed of the equivalent heat capacity of the cell-electrolyte and the electrolytic cell time constant.

[0111] Specifically, the ratio of the electrolytic cell time constant of 419.7s to the steady-state thermal gain of 0.00480℃ / kW is determined as the equivalent heat capacity of the cell-electrolyte, which is 87437.5kJ / K, or 87.4MJ / K. The thermal inertia characteristic of the cell-electrolyte is composed of the equivalent heat capacity of the cell-electrolyte of 87.4MJ / K and the electrolytic cell time constant of 419.7s.

[0112] The equivalent heat capacity of the tank-electrolyte reflects the heat absorbed or released by the tank and circulating electrolyte when the pipeline temperature changes by a unit. The time constant of the electrolytic cell reflects the rate at which the pipeline temperature approaches a new steady state after a change in net heat input.

[0113] S4. Load the candidate temperature control parameter combinations into a simulation model that connects the heat transfer execution model, the electrolytic cell thermal response model, and the installed temperature measurement heat transfer model in series. Select the combination with the smallest temperature deviation from the combinations that meet the temperature and heat transfer fluctuation constraints. Note that the following should be noted in this step: S4.1. The larger value between the pure time delay and the sampling interval is determined as the lower limit of the closed-loop time constant, and the sum of the lower limit of the closed-loop time constant and the electrolytic cell time constant is determined as the upper limit of the closed-loop time constant. The closed-loop time constant is selected between the lower limit and the upper limit of the closed-loop time constant according to the sampling interval. The closed-loop time constant, steady-state thermal gain, electrolytic cell time constant and pure time delay are substituted into the internal model control tuning relationship of the first-order inertial pure time delay model to obtain the candidate temperature control parameter combination composed of proportional coefficient, integral time and derivative time.

[0114] Specifically, the first-order inertial pure time delay model adopts the electrolytic cell thermal response model obtained in step S3. Its input is the net heat input increment, and its output is the pipeline side temperature rise. The model parameters include steady-state thermal gain of 0.00480℃ / kW, electrolytic cell time constant of 419.7s, and pure time delay duration of 12s.

[0115] The larger of the pure time delay of 12s and the sampling interval of 1s, 12s, is determined as the lower limit of the closed-loop time constant. The lower limit of the closed-loop time constant of 12s is added to the electrolytic cell time constant of 419.7s, and rounded up according to the 1s sampling interval to obtain the upper limit of the closed-loop time constant of 432s. The closed-loop time constant is selected sequentially in 1s intervals between 12s and 432s to form 421 closed-loop time constant values.

[0116] For each closed-loop time constant, the proportional gain, integral time, and derivative time are calculated according to the internal model control tuning relationship of the first-order inertial pure time delay model. The thermal domain value of the proportional gain is obtained according to the following relationship: first calculate the sum of twice the electrolytic cell time constant and the pure time delay, then calculate the sum of twice the closed-loop time constant and the pure time delay, multiply the steady-state thermal gain by the latter, and finally divide the former by the product. The integral time is the sum of the electrolytic cell time constant and 0.5 times the pure time delay. The derivative time is the product of the electrolytic cell time constant and the pure time delay, and then divided by the sum of twice the electrolytic cell time constant and the pure time delay.

[0117] The heat domain value of the above proportional coefficient corresponds to the conversion relationship between temperature deviation and heat exchange adjustment value, while the output of the temperature controller transmits the opening of the heat exchange regulating valve to the heat exchange execution model; the control station obtains the heat exchange gain in the valve opening domain based on the change in heat transfer amount corresponding to the unit valve opening change near the current valve working point, and then divides the heat domain proportional coefficient by the valve opening domain heat exchange gain to obtain the valve opening domain proportional coefficient.

[0118] The temperature controller uses a reverse adjustment direction; when the sensor temperature is higher than the target temperature, the temperature controller increases the opening of the heat exchange regulating valve; when the sensor temperature is lower than the target temperature, the temperature controller decreases the opening of the heat exchange regulating valve.

[0119] In this embodiment, when the closed-loop time constant is 120s, the heat domain proportionality coefficient is 703.9kW / ℃, the integral time is 425.7s, and the derivative time is 5.9s, based on the internal model control tuning relationship. Near the current valve operating point, for every 1% increase in the opening of the heat exchange regulating valve, the heat transfer amount increases by 21.6kW, thus obtaining a valve opening domain proportionality coefficient of 32.6 percentage points / ℃. A set of candidate temperature control parameter combinations is formed by the proportionality coefficient of 32.6 percentage points / ℃, the integral time of 425.7s, and the derivative time of 5.9s.

[0120] S4.2 Determine the cooling water mass flow rate based on the valve flow characteristics between the opening of the heat exchange regulating valve and the mass flow rate of the cooling water; calculate the cooling water heat capacity flow rate and the electrolyte heat capacity flow rate based on the cooling water mass flow rate, the circulating electrolyte mass flow rate, and the specific heat capacity of the media on both sides; calculate the heat exchanger effectiveness according to the relationship between effectiveness and number of heat transfer units based on the overall heat transfer coefficient, heat exchange area, and heat capacity flow rates on both sides; multiply the smaller value of the heat exchanger effectiveness and the heat capacity flow rates on both sides by the difference between the electrolyte inlet temperature and the cooling water inlet temperature to obtain the heat transfer outflow, thus forming the heat exchange execution model.

[0121] Furthermore, the valve flow characteristics between the opening of the heat exchange regulating valve and the mass flow rate of the cooling water were obtained based on the commissioning test of the heat exchange regulating valve. Under the conditions that the difference between the cooling water inlet pressure and the heat exchanger outlet pressure is 0.18 MPa and the cooling water temperature is 30℃~35℃, the opening of the heat exchange regulating valve was adjusted to 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% and 85% in sequence.

[0122] For each heat exchange regulating valve opening, if the difference between the maximum and minimum values ​​of the cooling water mass flow rate over 30 consecutive seconds is not greater than 0.5% of the average cooling water mass flow rate over 30 consecutive seconds, calculate the average cooling water mass flow rate over 30 consecutive seconds, and correlate the heat exchange regulating valve opening with the average cooling water mass flow rate to form the valve flow characteristics.

[0123] In the valve flow characteristics obtained in this embodiment, when the opening of the heat exchange regulating valve is 20%, 30%, 40%, 50%, 60%, 70%, and 80%, the mass flow rates of cooling water are 10.1 kg / s, 17.4 kg / s, 25.2 kg / s, 33.1 kg / s, 41.0 kg / s, 48.6 kg / s, and 55.2 kg / s, respectively.

[0124] When the opening of the heat exchange regulating valve in the simulation calculation is between two adjacent test openings, the mass flow rates of the two adjacent cooling waters are linearly converted according to the position of the current heat exchange regulating valve opening between the two adjacent test openings. When the opening of the heat exchange regulating valve is less than 5%, the mass flow rate of the cooling water is taken as the value corresponding to the 5% opening. When the opening of the heat exchange regulating valve is greater than 85%, the mass flow rate of the cooling water is taken as the value corresponding to the 85% opening.

[0125] The cooling water heat capacity flow rate is obtained by multiplying the cooling water mass flow rate by the cooling water specific heat capacity, and the electrolyte heat capacity flow rate is obtained by multiplying the circulating electrolyte mass flow rate by the electrolyte specific heat capacity. In this embodiment, when the heat exchange regulating valve opening is 40%, the cooling water mass flow rate is 25.2 kg / s, the cooling water specific heat capacity is 4.18 kJ / (kg·K), and the cooling water heat capacity flow rate is 105.3 kW / K; the circulating electrolyte mass flow rate is 33.0 kg / s, the electrolyte specific heat capacity is 3.75 kJ / (kg·K), and the electrolyte heat capacity flow rate is 123.8 kW / K.

[0126] The overall heat transfer coefficient of the heat exchanger is calculated based on the heat removal rate, the inlet and outlet temperatures of the medium on both sides, and the heat transfer area during the heat exchanger commissioning test. In this embodiment, it is taken as 0.32 kW / (m²). 2 ·K), heat exchange area is taken as 80m² 2 .

[0127] The control station selects the smaller and larger heat capacity flow rates from the cooling water heat capacity flow rate and the electrolyte heat capacity flow rate, respectively, and uses the ratio of the smaller heat capacity flow rate to the larger heat capacity flow rate as the heat capacity flow rate ratio on both sides. The total heat transfer coefficient of the heat exchanger is multiplied by the heat transfer area and then divided by the smaller heat capacity flow rate to obtain the number of heat transfer units. Subsequently, according to the correspondence between the effectiveness of the counter-flow heat exchanger and the number of heat transfer units, the effectiveness of the heat exchanger is obtained based on the number of heat transfer units and the heat capacity flow rate ratio on both sides.

[0128] Under the above-mentioned 40% heat exchange regulating valve opening condition, the minimum heat capacity flow rate is 105.3 kW / K, the maximum heat capacity flow rate is 123.8 kW / K, the ratio of heat capacity flow rates on both sides is 0.851, the number of heat transfer units is 0.243, and the corresponding heat exchanger effectiveness is 0.198. When the electrolyte inlet temperature is 82.5℃ and the cooling water inlet temperature is 31.0℃, multiplying the heat exchanger effectiveness of 0.198, the minimum heat capacity flow rate of 105.3 kW / K, and the temperature difference between the two inlets of 51.5K, the heat transfer output is obtained as 1075.7 kW.

[0129] Based on the above conversion relationship between valve opening and cooling water mass flow rate, the calculation relationship of heat capacity flow rate on both sides, and the processing relationship of effectiveness-number of heat transfer units, a heat exchange execution model is formed. The heat exchange execution model takes the opening of the heat exchange regulating valve, the mass flow rate of the circulating electrolyte, the electrolyte inlet temperature, and the cooling water inlet temperature as inputs, and the heat exchange removal amount as output.

[0130] S4.3 Connect the output of the temperature controller to the heat exchange regulating valve opening input of the heat exchange execution model. Input the net heat obtained by subtracting the heat transfer amount output by the heat exchange execution model from the electrochemical heat generation into the electrolytic cell thermal response model. Input the pipeline side temperature output by the electrolytic cell thermal response model into the installation temperature measurement heat transfer model. Feed back the sensor temperature output by the installation temperature measurement heat transfer model to the temperature controller to form a simulation model.

[0131] Specifically, the temperature controller calculates the heat exchange adjustment command based on the deviation between the target temperature and the sensor temperature, and converts the heat exchange adjustment command into the change in the opening of the heat exchange regulating valve based on the heat exchange gain of the valve opening domain corresponding to the current valve operating point; the change in the opening of the heat exchange regulating valve is added to the opening of the heat exchange regulating valve at the previous sampling time to obtain the opening of the heat exchange regulating valve at the current sampling time.

[0132] At each simulation sampling moment, calculations are performed in the following order: First, the temperature deviation is obtained based on the current sensor temperature and the target temperature; then, the opening degree of the heat exchange regulating valve is obtained based on the current candidate temperature control parameter combination; subsequently, the heat exchange execution model obtains the heat transfer amount based on the heat exchange regulating valve opening, cooling water inlet temperature, circulating electrolyte mass flow rate, and electrolyte inlet temperature; then, the net heat input is obtained by subtracting the heat transfer amount from the electrochemical heat generation, and the net heat input increment is input into the electrolytic cell thermal response model to obtain the pipeline side temperature; finally, the pipeline side temperature and ambient temperature are input into the installed temperature measurement heat transfer model to obtain the sensor temperature at the next sampling moment, and the sensor temperature is fed back to the temperature controller.

[0133] When the opening of the heat exchange regulating valve reaches the upper limit and the temperature deviation requirement continues to increase, the temperature controller stops accumulating the integral term in the direction of increasing the opening of the heat exchange regulating valve; when the opening of the heat exchange regulating valve reaches the lower limit and the temperature deviation requirement continues to decrease, the temperature controller stops accumulating the integral term in the direction of decreasing the opening of the heat exchange regulating valve.

[0134] S4.4 Configure the candidate temperature control parameter combinations as the control parameters of the temperature controller. Under the same electrolytic cell DC current change sequence, electrolytic cell total voltage change sequence, circulating electrolyte mass flow rate sequence, target temperature, initial pipeline side temperature, initial sensor temperature, cooling water inlet temperature, ambient temperature, simulation duration, and heat exchange regulating valve opening range, calculate the sensor temperature response and heat exchange regulating valve opening response.

[0135] Specifically, each candidate temperature control parameter combination is configured to the same temperature controller model; before each simulation begins, the integral term of the temperature controller is restored to 0, the historical input of the differential term is set to the initial temperature deviation, and the internal states of the electrolytic cell thermal response model and the installed temperature measurement heat transfer model are restored to the same initial state, so that there are only differences in control parameters between different candidate temperature control parameter combinations.

[0136] The simulation duration of this embodiment is 1800s, and the simulation sampling interval is 1s. From 0s to 299s, the DC current of the electrolytic cell is 5600A, the total voltage of the electrolytic cell is 730V, and the mass flow rate of the circulating electrolyte is 31.0kg / s. From 300s to 899s, the DC current of the electrolytic cell is 6200A, the total voltage of the electrolytic cell is 760V, and the mass flow rate of the circulating electrolyte is 33.0kg / s. From 900s to 1199s, the DC current of the electrolytic cell is 5800A, the total voltage of the electrolytic cell is 742V, and the mass flow rate of the circulating electrolyte is 32.0kg / s. From 1200s to 1799s, the DC current of the electrolytic cell is 6400A, the total voltage of the electrolytic cell is 775V, and the mass flow rate of the circulating electrolyte is 34.0kg / s.

[0137] All candidate temperature control parameter combinations use the same target temperature of 80.0℃, initial pipeline side temperature of 79.2℃, initial sensor temperature of 78.8℃, cooling water inlet temperature of 31.0℃, ambient temperature of 29.0℃, simulation duration of 1800s, initial heat exchange regulating valve opening of 40.0%, lower limit of heat exchange regulating valve opening of 5.0%, and upper limit of heat exchange regulating valve opening of 85.0%.

[0138] The above-mentioned DC current change sequence and total voltage change sequence of the electrolyzer correspond to the operation process of renewable energy input power increasing, decreasing and increasing again, while the circulating electrolyte mass flow rate sequence corresponds to the operating point of the circulating pump under different electrolysis loads.

[0139] At each sampling moment, the control station obtains the electrochemical heat generation based on the current DC current of the electrolytic cell, the total voltage of the electrolytic cell, and the total thermal neutral voltage of the electrolytic cell. It also obtains the heat transfer amount based on the opening of the heat exchange regulating valve and the state of the media on both sides. Then, it obtains the pipeline side temperature and sensor temperature sequentially through the electrolytic cell thermal response model and the installation temperature measurement heat transfer model. The sensor temperature and the opening of the heat exchange regulating valve are arranged according to the sampling moment to form the sensor temperature response and the heat exchange regulating valve opening response, respectively.

[0140] S4.5. Eliminate candidate temperature control parameter combinations that do not enter the target temperature allowable range for three consecutive sampling times within the simulation duration; take the response segment after the sensor temperature response of the remaining candidate temperature control parameter combinations enters the target temperature allowable range for the first three consecutive sampling times as the stability evaluation segment, calculate the difference between the maximum and minimum sensor temperature within the stability evaluation segment, and count the number of sign changes of adjacent non-zero heat exchange regulating valve opening changes; take the maximum values ​​of the two calculation results of the normal temperature control sample under the same simulation boundary conditions as the upper limit of temperature fluctuation and the upper limit of heat exchange fluctuation, respectively, and eliminate candidate temperature control parameter combinations whose sensor temperature exceeds the upper limit of electrolytic cell operating temperature or whose calculation result exceeds the corresponding upper limit.

[0141] Specifically, the target temperature range is determined based on the technical conditions for controlling the outlet temperature of the electrolytic cell. The allowable deviation for the outlet temperature control of the electrolytic cell is 0.5℃ above or below the target temperature. Therefore, when the target temperature is 80.0℃, the target temperature range is 79.5℃ to 80.5℃.

[0142] When the target temperature is adjusted, the adjusted target temperature is subtracted by 0.5℃ to obtain the lower limit of the target temperature allowable range, and the adjusted target temperature is added by 0.5℃ to obtain the upper limit of the target temperature allowable range. If the upper limit of the target temperature allowable range exceeds the upper limit of the electrolytic cell operating temperature, the upper limit of the electrolytic cell operating temperature shall be used as the upper limit of the target temperature allowable range.

[0143] The control station checks the sensor temperature response from the simulation start point; when the sensor temperature is between 79.5℃ and 80.5℃ for three consecutive sampling times, the first sampling time of the three consecutive sampling times is determined as the start time of the stable evaluation segment, and the response range between the start time of the stable evaluation segment and the simulation end point is determined as the stable evaluation segment; candidate temperature control parameter combinations that do not have three consecutive sampling times within the target temperature allowable range are eliminated during the simulation time.

[0144] For each stable evaluation segment, the maximum and minimum sensor temperatures are selected, and the difference between them is calculated. The control station also calculates the change in the opening of the heat exchange regulating valve at adjacent sampling times. The calculation resolution of the heat exchange regulating valve opening is 0.1 percentage points, and the change in absolute value less than 0.1 percentage points is treated as 0. When the positive and negative signs of two adjacent non-zero heat exchange regulating valve opening changes are different, the direction of the heat exchange regulating valve opening change is recorded as a change of 1.

[0145] Under the same conditions of DC current change sequence, total voltage change sequence, circulating electrolyte mass flow rate sequence, target temperature, cooling water inlet temperature, and ambient temperature, 20 groups of normal temperature control samples with stable operation and no heat exchange regulating valve jamming were selected. The maximum difference between the maximum and minimum temperature values ​​in the stable evaluation segment of the normal temperature control samples was 0.86℃, and the maximum number of times the opening direction of the heat exchange regulating valve changed was 12. Therefore, 0.86℃ was determined as the upper limit of temperature fluctuation, and 12 times was determined as the upper limit of heat exchange fluctuation.

[0146] If the sensor temperature of any candidate temperature control parameter combination exceeds 85.0℃, the difference between the maximum and minimum temperature values ​​in the stable evaluation section exceeds 0.86℃, or the opening direction of the heat exchange regulating valve changes more than 12 times, the candidate temperature control parameter combination will be removed.

[0147] Of the 421 candidate temperature control parameter combinations in this embodiment, 97 combinations did not enter the target temperature allowable range within 1800s, 41 combinations had sensor temperatures exceeding 85.0℃, 238 combinations had temperature fluctuations or heat exchange fluctuations exceeding the corresponding upper limit, leaving 45 candidate temperature control parameter combinations.

[0148] S4.6 When there are no remaining candidate temperature control parameter combinations, the closed-loop time constant value is increased by one sampling interval to form new candidate temperature control parameter combinations. Under the same simulation boundary conditions, the sensor temperature response and heat exchange regulating valve opening response are recalculated, and combinations that do not enter the target temperature allowable range are eliminated and a stable evaluation segment is selected. When there are remaining candidate temperature control parameter combinations, the cumulative absolute deviation between the sensor temperature response and the target temperature is calculated within the same simulation duration. The candidate temperature control parameter combination with the smallest cumulative absolute deviation is selected as the target temperature control parameter combination. When the cumulative absolute deviation values ​​are the same, the candidate temperature control parameter combination with the earlier start time of the stable evaluation segment is selected as the target temperature control parameter combination.

[0149] Specifically, when there are no remaining candidate temperature control parameter combinations, the original lower limit of the closed-loop time constant and the original upper limit of the closed-loop time constant are increased by 1s respectively, and a new candidate temperature control parameter combination is formed between the increased lower limit of the closed-loop time constant and the upper limit of the closed-loop time constant at a 1s interval. Then, steps S4.4 and S4.5 are repeated according to the same simulation boundary conditions.

[0150] If no remaining candidate temperature control parameter combinations are found after the overall closed-loop time constant has increased 20 times, the control parameters are stopped from being sent to the temperature controller, and abnormal results of the model or heat exchange execution conditions are output. Abnormal results include the effective opening range of the heat exchange regulating valve, the cooling water inlet temperature, the total heat transfer coefficient of the heat exchanger, the installation contact status of the outlet temperature sensor, and the fitting residual of the electrolytic cell thermal response model. After the equipment is checked, steps S1 to S4 are re-executed.

[0151] When there are remaining candidate temperature control parameter combinations, the control station obtains the absolute difference between the sensor temperature and the target temperature for each remaining candidate temperature control parameter combination from the simulation start point to the simulation end point. Each absolute difference is multiplied by a 1-second sampling interval and then accumulated to obtain the cumulative absolute deviation value.

[0152] In this embodiment, the candidate temperature control parameter combinations with closed-loop time constants of 119s, 120s, and 121s all meet the temperature and heat transfer fluctuation constraints, and their cumulative absolute deviation values ​​are 331.8℃·s, 326.4℃·s, and 326.4℃·s, respectively. The stable evaluation segment start time corresponding to the candidate temperature control parameter combination with a closed-loop time constant of 120s is 214s, and the stable evaluation segment start time corresponding to the candidate temperature control parameter combination with a closed-loop time constant of 121s is 219s. Therefore, the candidate temperature control parameter combination with a closed-loop time constant of 120s is selected as the target temperature control parameter combination.

[0153] Preferably, this embodiment solves the problem that it is difficult to simultaneously consider the upper limit of the electrolytic cell operating temperature, the delay in the installation of the outlet temperature sensor, and the frequent switching of the heat exchange regulating valve when adjusting the temperature control parameters based solely on experience. By first eliminating candidate temperature control parameter combinations that have not entered the target temperature allowable range, exceed the upper limit of the electrolytic cell operating temperature, or exceed the fluctuation constraint, and then selecting based on the cumulative absolute deviation value and the start time of the stability evaluation segment, the resulting target temperature control parameter combination simultaneously considers temperature deviation, response speed, and heat exchange execution stability.

[0154] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for generating temperature control parameters based on thermal inertia characteristics, characterized in that, include: When the DC current of the electrolytic cell, the flow rate of the circulating electrolyte, and the opening degree of the heat exchange regulating valve are constant, and the reference temperature on the pipeline side is in a stable range, a pulse power with the same amplitude and duration as the reference contact sample is applied to the resistive heating element attached to the pipeline side of the temperature sensor mounting interface; the pipeline side reference temperature is determined according to the ratio of voltage to current across the resistive heating element and the resistance temperature characteristics, forming a pipeline side reference temperature response, and a sensor temperature response is formed through the temperature sensor; The reference temperature fluctuation range and the sensor temperature fluctuation range are respectively defined by the minimum and maximum values ​​of the pipeline-side reference temperature and the minimum and maximum values ​​of the sensor temperature at three consecutive sampling times before excitation. The first sampling moment when the pipeline-side reference temperature is higher than the upper limit of the reference temperature fluctuation range for two consecutive sampling moments is determined as the reference response start point, and the first sampling moment when the sensor temperature is higher than the upper limit of the sensor temperature fluctuation range for two consecutive sampling moments is determined as the sensor response start point. Calculate the response delay between the reference response start point and the sensor response start point; using the average reference temperature on the pipeline side before excitation and the average sensor temperature as benchmarks, calculate the reference temperature rise peak and the sensor temperature rise peak, and determine the normalized amplitude attenuation by subtracting the ratio of the sensor temperature rise peak to the reference temperature rise peak. The contact response characteristics are composed of the response delay and the normalized amplitude attenuation; The method for calculating the response delay includes: subtracting the sampling number of the reference response start point from the sampling number of the sensor response start point to obtain the sampling number difference, and determining the response delay by multiplying the sampling number difference by the sampling interval. Using the pipeline-side reference temperature and circulating electrolyte flow rate before local thermal excitation as the matching temperature and matching flow rate, respectively, reference contact response features with the same temperature and flow rate as the matching temperature and matching flow rate are selected from the closely contacted samples. When the reference contact response features do not exist, four sets of adjacent reference contact response features with temperature and flow rate around the matching temperature and matching flow rate are selected, and reference contact response features with the same temperature and flow rate are obtained by bilinear interpolation. The response delay deviation is obtained by subtracting the response delay in the reference contact response feature from the response delay in the contact response feature; the amplitude attenuation deviation is obtained by subtracting the normalized amplitude attenuation in the reference contact response feature from the normalized amplitude attenuation in the contact response feature. Calculate the median of response delay deviation and the median of amplitude attenuation deviation for closely contacted samples, samples with increased contact thermal resistance, and samples with partial detachment, respectively; determine the midpoint of the median of response delay deviation between closely contacted samples and samples with increased contact thermal resistance as the first boundary of response delay, and determine the midpoint of the median of response delay deviation between samples with increased contact thermal resistance and samples with partial detachment as the second boundary of response delay; determine the first boundary and the second boundary of amplitude attenuation in the same way. The contact state levels to which the response delay deviation and the amplitude attenuation deviation belong are determined respectively, and the higher of the two contact state levels is determined as the installation contact state of the temperature sensor in the order of close contact level, contact thermal resistance increase level and partial separation level from low to high. Substitute the pipeline-side reference temperature response, sensor temperature response, and ambient temperature of the sample to which the installation contact state belongs into the heat balance equation of the sensor thermal node. Under the constraint that the installation contact thermal resistance, sensor heat dissipation thermal resistance, and sensor heat capacity are all greater than zero, nonlinear least squares identification is performed with the condition that the sum of the squared deviations between the calculated sensor temperature value and the measured sensor temperature value is minimized. The median values ​​of the identified installation contact thermal resistance, sensor heat dissipation thermal resistance, and sensor thermal capacity are taken respectively, and the installation contact thermal resistance, the sensor heat dissipation thermal resistance, and the sensor thermal capacity are loaded into a two-node thermal resistance-thermal capacity model to obtain the installation temperature measurement heat transfer model. The two-node thermal resistance-thermal capacity model includes a pipeline-side thermal node and a sensor thermal node. The pipeline-side thermal node and the sensor thermal node are connected through the mounting contact thermal resistance. The sensor thermal node is associated with the sensor thermal capacity and is connected to the ambient temperature boundary through the sensor heat dissipation thermal resistance. The installation temperature measurement heat transfer model takes the pipeline-side temperature and ambient temperature as inputs and the sensor temperature as output. The sensor temperature response when the net heat input changes is inverted using the aforementioned installation temperature measurement heat transfer model to obtain the pipeline-side temperature response; the electrolytic cell thermal response model is identified based on the net heat input and pipeline-side temperature response, and the thermal inertia characteristics of the cell-electrolyte are extracted. The candidate temperature control parameter combinations are loaded into a simulation model that connects the heat exchange execution model, the electrolytic cell thermal response model, and the installation temperature measurement heat transfer model. The combination with the smallest temperature deviation is selected from the combinations that meet the temperature and heat exchange fluctuation constraints.

2. The method for generating temperature control parameters based on thermal inertia characteristics according to claim 1, characterized in that, The determination of the contact state level includes: When the response delay deviation is lower than the first response delay boundary, it is determined to be a close contact level; when the response delay deviation reaches the first response delay boundary but is lower than the second response delay boundary, it is determined to be a contact thermal resistance increase level; when the response delay deviation reaches the second response delay boundary, it is determined to be a partial detachment level. When the amplitude attenuation deviation is lower than the first amplitude attenuation boundary, it is determined to be the tight contact level; when the amplitude attenuation deviation reaches the first amplitude attenuation boundary but is lower than the second amplitude attenuation boundary, it is determined to be the increased contact thermal resistance level; when the amplitude attenuation deviation reaches the second amplitude attenuation boundary, it is determined to be the partial detachment level.

3. The method for generating temperature control parameters based on thermal inertia characteristics according to claim 2, characterized in that, The obtained pipeline-side temperature response includes: The electrochemical heat generation is determined by multiplying the difference between the total voltage of the electrolytic cell and the thermal neutral voltage of the electrolytic cell with the DC current of the electrolytic cell; the heat transfer amount is determined by multiplying the mass flow rate of the circulating electrolyte, the specific heat capacity of the electrolyte, and the difference between the inlet and outlet electrolyte temperatures of the electrolyte heat exchanger; the net heat input is obtained by subtracting the heat transfer amount from the electrochemical heat generation. When the resistive heating element is not powered on, the sensor temperature response, ambient temperature and net heat input formed by the temperature sensor during the net heat input change are aligned according to the sampling time, and the sensor temperature change rate is obtained by dividing the difference of sensor temperature at adjacent sampling times by the sampling interval. The total heat flux of the sensor's thermal node is obtained by multiplying the sensor's heat capacity by the sensor's temperature change rate, adding the difference between the sensor's temperature and the ambient temperature by the ratio obtained by dividing the sensor's thermal resistance by its heat dissipation resistance; the total heat flux of the sensor's thermal node is then added to the product of the sensor's total heat flux and the installation contact thermal resistance, and the sensor's temperature is added to obtain the pipe-side temperature, and the pipe-side temperatures are arranged according to the sampling time to form the pipe-side temperature response.

4. The method for generating temperature control parameters based on thermal inertia characteristics according to claim 3, characterized in that, The thermal inertia characteristics of the extraction tank-electrolyte include: Based on the net heat input and pipe-side temperature at the identification starting point, net heat input increment and pipe-side temperature rise are formed respectively. Candidate pure lag durations are set with different integer multiple sampling intervals. For each candidate pure lag duration, the pipe-side temperature rise at the current sampling time is used as the dependent variable, and the pipe-side temperature rise corresponding to the previous sampling time adjacent to the current sampling time and the net heat input increment corresponding to the sampling time after tracing back the candidate pure lag duration from the current sampling time are used as independent variables. Least square fitting is performed under the condition that the discrete attenuation coefficient is greater than zero and less than one, and the heat input response coefficient is non-negative, to obtain the discrete attenuation coefficient, heat input response coefficient and fitting residual. The candidate pure lag time with the smallest fitting residual is determined as the pure lag time. The sampling interval is divided by the absolute value of the natural logarithm of the discrete attenuation coefficient to obtain the electrolytic cell time constant. The thermal input response coefficient is divided by the difference obtained by subtracting the discrete attenuation coefficient to obtain the steady-state thermal gain. The electrolytic cell thermal response model is composed of the electrolytic cell time constant, the steady-state thermal gain, and the pure lag time. The ratio of the electrolytic cell time constant to the steady-state thermal gain is determined as the equivalent heat capacity of the cell-electrolyte. The equivalent heat capacity of the cell-electrolyte and the electrolytic cell time constant constitute the thermal inertia characteristic of the cell-electrolyte.

5. The method for generating temperature control parameters based on thermal inertia characteristics according to claim 4, characterized in that, The formation of the simulation model includes: The larger value between the pure time delay duration and the sampling interval is determined as the lower limit of the closed-loop time constant, and the sum of the lower limit of the closed-loop time constant and the electrolytic cell time constant is determined as the upper limit of the closed-loop time constant. The closed-loop time constant is selected between the lower limit of the closed-loop time constant and the upper limit of the closed-loop time constant according to the sampling interval. The closed-loop time constant, the steady-state thermal gain, the electrolytic cell time constant and the pure time delay duration are substituted into the internal model control tuning relationship of the first-order inertial pure time delay model to obtain a candidate temperature control parameter combination composed of proportional coefficient, integral time and derivative time. The cooling water mass flow rate is determined based on the valve flow characteristics between the opening of the heat exchange regulating valve and the mass flow rate of the cooling water. The cooling water heat capacity flow rate and the electrolyte heat capacity flow rate are calculated based on the cooling water mass flow rate, the circulating electrolyte mass flow rate, and the specific heat capacity of the media on both sides. The heat exchanger effectiveness is calculated based on the overall heat transfer coefficient, heat exchange area, and heat capacity flow rates on both sides, according to the relationship between effectiveness and number of heat transfer units. The smaller value of the heat exchanger effectiveness and the heat capacity flow rates on both sides is multiplied by the difference between the electrolyte inlet temperature and the cooling water inlet temperature to obtain the heat removal amount, thus forming the heat exchange execution model. Connect the output of the temperature controller to the heat exchange regulating valve opening input of the heat exchange execution model, input the net heat obtained by subtracting the heat transfer outflow output by the heat exchange execution model from the electrochemical heat generation into the electrolytic cell thermal response model, input the pipeline side temperature output by the electrolytic cell thermal response model into the installation temperature measurement heat transfer model, and feed back the sensor temperature output by the installation temperature measurement heat transfer model to the temperature controller to form the simulation model.

6. The method for generating temperature control parameters based on thermal inertia characteristics according to claim 5, characterized in that, The selection of the combination with the smallest temperature deviation includes: The candidate temperature control parameters are configured as the control parameters of the temperature controller. Under the same electrolytic cell DC current change sequence, electrolytic cell total voltage change sequence, circulating electrolyte mass flow rate sequence, target temperature, initial pipeline side temperature, initial sensor temperature, cooling water inlet temperature, ambient temperature, simulation duration, and heat exchange regulating valve opening range, the sensor temperature response and heat exchange regulating valve opening response are calculated. Candidate temperature control parameter combinations that do not enter the target temperature allowable range for three consecutive sampling times within the simulation duration are eliminated. The response segment after the sensor temperature response of the remaining candidate temperature control parameter combinations enters the target temperature allowable range for the first three consecutive sampling times is taken as the stability evaluation segment. The difference between the maximum and minimum sensor temperature within the stability evaluation segment is calculated, and the number of sign changes of adjacent non-zero heat exchange regulating valve opening changes is counted. The maximum values ​​of the two calculation results of the normal temperature control sample under the same simulation boundary conditions are taken as the upper limit of temperature fluctuation and the upper limit of heat exchange fluctuation, respectively. Candidate temperature control parameter combinations whose sensor temperature exceeds the upper limit of electrolytic cell operating temperature or whose calculation result exceeds the corresponding upper limit are eliminated. If no remaining candidate temperature control parameter combinations exist, the closed-loop time constant value is increased by one sampling interval to re-form candidate temperature control parameter combinations. The sensor temperature response and heat exchange regulating valve opening response are recalculated under the same simulation boundary conditions. Combinations that do not enter the target temperature allowable range are eliminated, and a stable evaluation segment is selected. If remaining candidate temperature control parameter combinations exist, the cumulative absolute deviation between the sensor temperature response and the target temperature is calculated within the same simulation duration. The candidate temperature control parameter combination with the smallest cumulative absolute deviation value is selected as the target temperature control parameter combination. If the cumulative absolute deviation values ​​are the same, the candidate temperature control parameter combination with the earlier start time of the stable evaluation segment is selected as the target temperature control parameter combination.

Images