CN-121992450-A - Method and system for monitoring and dispatching optimization of running state of composite membrane electrolytic tank

CN121992450ACN 121992450 ACN121992450 ACN 121992450ACN-121992450-A

Abstract

The invention discloses a method and a system for monitoring and dispatching optimization of the running state of a composite membrane electrolytic tank, and particularly relates to the technical field of hydrogen production by water electrolysis; the method comprises the steps of determining a health state and a corresponding life calibration coefficient through piecewise linear mapping based on a comprehensive health index, calculating a degradation rate by combining a time sequence health index to adjust a life optimization coefficient, simultaneously introducing the ratio of the average value of historical operating power to rated power, quantifying the nonlinear acceleration effect of power on aging, finally calculating the residual life through substituting the obtained estimated reduction coefficient into a formula, fully fusing the real-time state and the power influence of equipment, and obviously reducing the deviation between a predicted result and the actual life.

Inventors

WANG ZIHAO
WANG PENG
LIU LIN

Assignees

北京中电丰业技术开发有限公司

Dates

Publication Date: 20260508
Application Date: 20260324

Claims (9)

1. The method for monitoring and dispatching the operation state of the composite membrane electrolytic tank is characterized by comprising the following steps of: S1, constructing a multisource characteristic data set containing electrochemical impedance spectrum data, infrared thermal imaging data and sound emission data; s2, a deep learning algorithm is adopted to fuse the multisource characteristic data set to output a comprehensive health index, the comprehensive health index and the historical operating power are taken as inputs, and the estimated residual life of the equipment is output by combining the service life of the equipment; s3, dynamically calculating an absolute safe power upper limit and a recommended high-efficiency power interval of the electrolytic tank in the current state by combining the output comprehensive health index and the estimated residual life, and taking an intersection of the absolute safe power upper limit and the recommended high-efficiency power interval as a dynamic operation window; And S4, taking the total profit maximization in the scheduling period as an objective function, wherein the objective function explicitly comprises production income, energy consumption cost and aging cost, and performing rolling optimization solving by taking a dynamic operation window as a core constraint condition to obtain an optimal power instruction.
2. The method for monitoring and optimizing the operation state of the composite membrane electrolytic tank according to claim 1, which is characterized in that: s2, fusing the multisource characteristic data set to output a comprehensive health index; analyzing the electrochemical impedance spectrum data, and extracting a charge transfer resistor and a membrane resistor; Analyzing the infrared thermal imaging data, and extracting the highest temperature of a film area, the temperature difference between an anode and a cathode and the area ratio of hot spots; analyzing the acoustic emission data, and extracting the peak amplitude of the signal and the duty ratio of the high-frequency event; determining normal operation values corresponding to each type of data source respectively based on the historical normal data; The characteristic parameters extracted by each type of data source are combined with the corresponding normal operation values to be weighted and integrated, so that an electric health index, a thermal health index and a mechanical health index are obtained; and after normalizing the electric health index, the thermal health index and the mechanical health index, obtaining the comprehensive health index of the equipment by using weighted fusion treatment.
3. The method for monitoring and optimizing the operation state of the composite membrane electrolytic tank according to claim 2, which is characterized in that: s2, taking the comprehensive health index and the historical operating power as inputs; setting a piecewise linear mapping rule between the comprehensive health index and the health state, and converting the comprehensive health index into the health state, wherein the health state comprises a normal state, a moderate degradation state and a serious degradation state, and each group of health states corresponds to a service life calibration coefficient respectively; extracting comprehensive health indexes of x periods before the current time point, and recording the comprehensive health indexes as time sequence health indexes; Respectively marking the time sequence health index and the comprehensive health index of the current time point as According to the formula Calculating the degradation rate ; Rate of degradation Multiplying the obtained product by a preset rate sensitivity coefficient to obtain a trend adjustment coefficient; summing the accumulated value of the trend adjustment coefficient and the service life calibration coefficient with an integer one to serve as a service life optimization coefficient of the equipment; extracting historical running power of near x periods before a current time point, and taking the average value as average running power; Taking average running power as a numerator and rated power as a denominator to calculate a ratio, and recording as ; By passing through Calculating a power correction coefficient Wherein Is a preset power aging coefficient.
4. A method for monitoring and optimizing the operation state of a composite membrane electrolytic cell according to claim 3, wherein the method comprises the following steps: S2, outputting the estimated residual life of the equipment by combining the service life of the equipment; multiplying the life optimization coefficient by the power aging coefficient to obtain a pre-estimated reduction coefficient; And estimating the residual life by a formula = equipment theoretical life-used life x estimated reduction coefficient, and outputting the estimated residual life of the equipment.
5. The method for monitoring and optimizing the operation state of the composite membrane electrolytic tank according to claim 1, which is characterized in that: s3, dynamically calculating the absolute safe power upper limit of the electrolytic tank in the current state; calculating the ratio by taking the estimated residual life of the equipment as a numerator and the theoretical life of the equipment as a denominator to obtain the residual use index of the equipment; after carrying out normalization processing on the residual usage index and the comprehensive health index of the equipment, obtaining a state evaluation index of the equipment by utilizing weighted calculation logic processing; Setting a piecewise linear mapping rule between the state evaluation index and the safety adjustment coefficient by taking rated power as a reference, converting the state evaluation index into the safety adjustment coefficient, and multiplying the safety adjustment coefficient by the reference to obtain the absolute safety power upper limit of the electrolytic tank in the current state.
6. The method for monitoring and optimizing the operation state of the composite membrane electrolytic tank according to claim 1, which is characterized in that: S3, dynamically calculating a recommended high-efficiency power interval of the electrolytic tank in the current state; And identifying maximum power and minimum power meeting Q higher than a preset reference value in a constraint range by taking the upper limit of absolute safety power as a constraint, and taking a section of the maximum power and the minimum power as a recommended high-efficiency power section, wherein Q=product yield/energy consumption.
7. The method for monitoring and optimizing the operation state of the composite membrane electrolytic tank according to claim 1, which is characterized in that: s4, the objective function explicitly comprises production benefits, energy consumption cost and aging cost; Marking production benefits, energy consumption costs and aging costs as 、、 ; The variables are defined as follows: = Wherein For the yield at the time t, = ×v, The power at time t, v is the yield coefficient, Is the unit price of the product; ; In order to be able to take the length of the time step, Is electricity price; ; The aging cost coefficient is preset; Representing a lifetime calibration factor; Wherein the method comprises the steps of Defined within a dynamic operating window.
8. The method for monitoring and optimizing the operation state of the composite membrane electrolytic tank according to claim 7, which is characterized in that: S4, taking the dynamic operation window as a core constraint condition, and performing rolling optimization solution to obtain an optimal power instruction; The objective function formula is expressed as: wherein T is a scheduling period, and T is a time step in the period; to produce benefits, Is energy consumption cost, Is ageing cost; Outputting the optimal power obtained by solving as an optimal power instruction, and entering the next optimization period.
9. A system for monitoring and optimizing the operation state of a composite membrane electrolyzer, which is applied to the method for monitoring and optimizing the operation state of a composite membrane electrolyzer as recited in any one of the claims 1 to 8, and is characterized by comprising: The data input module is used for inputting electrochemical impedance spectrum data, infrared thermal imaging data and sound emission data and integrating the electrochemical impedance spectrum data, the infrared thermal imaging data and the sound emission data into a multi-source characteristic data set; The health evaluation module is used for fusing all the characteristics in the multi-source characteristic data set to output a comprehensive health index by adopting a deep learning algorithm, taking the comprehensive health index and the historical operating power as inputs, and outputting the estimated residual life of the equipment by combining the service life of the equipment; The boundary constraint module is used for dynamically calculating an absolute safe power upper limit and a recommended high-efficiency power interval of the electrolytic cell in the current state by combining the output comprehensive health index and the estimated residual life, and taking an intersection of the absolute safe power upper limit and the recommended high-efficiency power interval as a dynamic operation window; The constraint adjustment module is used for taking the total profit maximization in the scheduling period as an objective function, the objective function explicitly comprises production income, energy consumption cost and aging cost, and the dynamic operation window is used as a core constraint condition to perform rolling optimization solution so as to obtain an optimal power instruction.

Description

Method and system for monitoring and dispatching optimization of running state of composite membrane electrolytic tank Technical Field The invention relates to the technical field of hydrogen production by water electrolysis, in particular to a method and a system for monitoring and dispatching optimization of the running state of a composite membrane electrolytic tank. Background In the technical field of hydrogen production by water electrolysis, the composite membrane electrolytic tank is used as core production equipment, the running stability, the energy efficiency level and the service life of the composite membrane electrolytic tank directly determine the economical efficiency and the safety of a hydrogen production system, and the composite membrane electrolytic tank is gradually upgraded to a high-power long-period running direction along with the large-scale development of hydrogen energy industry. However, the existing operation monitoring and scheduling technology has the following defects in the actual application process: The traditional composite membrane electrolytic cell state monitoring depends on a single data source, and cannot comprehensively reflect the health state of equipment under the coupling of multiple physical fields, so that the deviation between the health evaluation and the actual equipment state is obvious, and buried potential safety hazards are scheduled for subsequent operation; On the one hand, the service life of the equipment is only used as a basic parameter, and the real-time health state is not combined for calibration, so that the deviation between a prediction result and the actual service life is larger; The prior art generally sets a fixed and conservative upper limit of safe power without considering the dynamic change of the health state in the whole life cycle of the equipment, when the equipment is in the early or middle stage of health, the fixed boundary limits the capacity potential of the equipment and cannot fully exert the high-efficiency operation advantages of the equipment, and when the equipment enters the health degradation stage, the original fixed boundary is difficult to match with the actual safety requirement, thereby possibly causing safety accidents such as thermal runaway, membrane breakage and the like, not only guaranteeing the production safety, but also reducing the flexibility and economy of the operation of the equipment. Therefore, a method and a system for monitoring and dispatching optimization of the running state of the composite membrane electrolytic tank are provided. Disclosure of Invention In order to overcome the defects in the prior art, the embodiment of the invention provides a method and a system for monitoring and dispatching optimization of the running state of a composite membrane electrolytic cell. In order to achieve the above purpose, the present invention provides the following technical solutions: a method for monitoring and dispatching and optimizing the running state of a composite membrane electrolytic tank comprises the following steps: S1, constructing a multisource characteristic data set containing electrochemical impedance spectrum data, infrared thermal imaging data and sound emission data; s2, a deep learning algorithm is adopted to fuse the multisource characteristic data set to output a comprehensive health index, the comprehensive health index and the historical operating power are taken as inputs, and the estimated residual life of the equipment is output by combining the service life of the equipment; s3, dynamically calculating an absolute safe power upper limit and a recommended high-efficiency power interval of the electrolytic tank in the current state by combining the output comprehensive health index and the estimated residual life, and taking an intersection of the absolute safe power upper limit and the recommended high-efficiency power interval as a dynamic operation window; And S4, taking the total profit maximization in the scheduling period as an objective function, wherein the objective function explicitly comprises production income, energy consumption cost and aging cost, and performing rolling optimization solving by taking a dynamic operation window as a core constraint condition to obtain an optimal power instruction. Specifically, in the step S2, a multisource characteristic data set is fused to output a comprehensive health index; analyzing the electrochemical impedance spectrum data, and extracting a charge transfer resistor and a membrane resistor; Analyzing the infrared thermal imaging data, and extracting the highest temperature of a film area, the temperature difference between an anode and a cathode and the area ratio of hot spots; analyzing the acoustic emission data, and extracting the peak amplitude of the signal and the duty ratio of the high-frequency event; determining normal operation values corresponding to each type of data source respectively based on the histo