Remote safety monitoring system for PEM water electrolysis hydrogen production equipment

Abstract

The invention discloses a remote safety monitoring system for PEM water electrolysis hydrogen production equipment, and particularly relates to the technical field of equipment monitoring. Electrical parameters during fault shutdown are extracted, working condition types during faults are recognized in combination with operation logs, matching of threshold values and working conditions is ensured, the electrical parameters before a single group of faults are divided into evaluation data packets, and the evaluation data packets are stored in a database; calculating a voltage deviation ratio and a current deviation ratio, constructing coordinate points, drawing a rectangle according to the coordinate points, taking the area of the rectangle as an electrical index, classifying according to working conditions to form an electrical threshold set of low, medium and full loads, reading the working conditions and electrical parameters of the current equipment in real time, calculating the current electrical index, and automatically calling the electrical threshold of the corresponding working condition. The early warning accuracy is greatly improved, and the problem that in the prior art, a single monitoring threshold value is set by depending on a fixed rated value is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device monitoring, more particularly, the present application relates to a remote safety monitoring system for a PEM water electrolysis hydrogen production device. BACKGROUND

[0002] With the transformation of global energy structure to clean energy, hydrogen energy as a clean, efficient and sustainable secondary energy has been widely concerned. PEM water electrolysis hydrogen production technology has become one of the important technologies for current hydrogen energy production due to its fast start-up speed, high current density, high hydrogen purity and environmental friendliness, and is widely used in energy, chemical industry, transportation and other fields. However, if the PEM water electrolysis hydrogen production equipment appears abnormal during operation and is not discovered and handled in time, it may cause equipment failure, even cause hydrogen leakage, fire, explosion and other serious safety accidents, resulting in personnel injury and property loss. At present, the monitoring system for PEM water electrolysis hydrogen production equipment still has the following shortcomings in practical application: Firstly, the electrical parameter monitoring method is rigid, and the early warning accuracy is insufficient. The existing technology mainly relies on fixed rated value to set a single monitoring threshold, and does not dynamically adjust in combination with historical fault data in the whole life cycle of the equipment. Since the normal range of electrical parameters of PEM water electrolysis hydrogen production equipment under different working conditions such as low load, medium load and full load is significantly different, a single threshold cannot adapt to multiple working condition scenes. In addition, the monitoring dimension of membrane state parameters is single, and the damage level judgment is missing. As the core component of PEM water electrolysis hydrogen production equipment, the state of proton exchange membrane directly determines the running efficiency and safety of the equipment. The existing monitoring system mainly judges a single parameter such as membrane humidity, membrane conductivity or membrane thickness, and does not realize the fusion analysis of multiple parameters, so it cannot accurately locate the degree of membrane damage and the corresponding risk level.

[0003] Therefore, a remote safety monitoring system for a PEM water electrolysis hydrogen production device is provided. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a remote safety monitoring system for a PEM water electrolysis hydrogen production device.

[0005] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme: A remote safety monitoring system for a PEM water electrolysis hydrogen production device comprises the following modules: A device end detection module: real-time acquisition of safety data of the PEM water electrolysis hydrogen production device, the safety data including electrical parameters, membrane state parameters and fluid parameters; An edge computing module: including a data storage unit and an intelligent analysis unit; The data storage unit receives safety data of the PEM electrolytic water hydrogen production equipment and stores electrical parameter historical data of the equipment, wherein the electrical parameter historical data is electrical parameter data of each group when a fault shutdown occurs; The intelligent analysis unit comprehensively analyzes the electrical threshold determined by the electrical parameter combined with the electrical parameter historical data after analyzing the safety data, and selectively triggers the early warning signaling to send to the cloud monitoring module; The cloud monitoring module selectively determines the early warning trigger reason by using the analysis algorithm logic after receiving the early warning signaling and identifying the trigger basis of the early warning signaling, and integrates and sends the early warning signaling to the mobile terminal of the operation and maintenance personnel.

[0006] Specifically, the specific process of determining the electrical threshold by using the electrical parameter historical data is as follows: The electrical parameters include the electrolytic cell voltage and the electrolytic cell current; Identify the working condition type of the equipment when each group of faults occurs, including low load, medium load, and full load; Divide the electrical parameters of the equipment when a single group of faults occurs into the electrical parameters of the running time zone before the fault occurs according to the time axis, as the evaluation data package; For the evaluation data package, the average values of the electrolytic cell voltage and the electrolytic cell current at each time point in the running time zone are calculated as the total voltage and the total current; Calculate the deviation rate between the total voltage, the total current, and the corresponding rated total voltage and rated total current to determine the voltage deviation rate and the current deviation rate; After representing the voltage deviation rate and the current deviation rate as a and b respectively, a coordinate point (a, b) is constructed, a plane rectangular coordinate system is drawn, the coordinate point (a, b) is drawn in the plane rectangular coordinate system, and a vertical line is drawn from the coordinate point (a, b) to the X-axis and Y-axis directions respectively until the two vertical lines intersect with the X-axis and Y-axis respectively. The area of the figure enclosed by the two vertical lines and the X-axis and Y-axis is calculated as the electrical index; After the electrical parameter data of each group of faults is calculated by the electrical index, the electrical index of each group is classified according to the working condition type of the equipment when the fault occurs, as the low load index set, the medium load index set, and the full load index set; Each group of electrical indexes in the low load index set, the medium load index set, and the full load index set is used as the electrical threshold.

[0007] Specifically, the specific process of selectively triggering the early warning signaling is as follows: The electrical parameter data and the working condition type of the PEM electrolytic water hydrogen production equipment in the current time zone are read, the electrical condition index of the PEM electrolytic water hydrogen production equipment in the current time zone is calculated, the electrical threshold values of each group corresponding to the working condition are automatically called, the electrical condition index calculated in the current time zone is compared with the electrical threshold values of each group corresponding to the working condition, and if the electrical condition index in the current time zone is higher than one of the electrical threshold values, a pre-warning signaling is triggered.

[0008] Specifically, after the trigger of the identification pre-warning signaling, the pre-warning trigger reason is determined by selectively using the analysis algorithm logic: The number of electrical threshold values lower than the electrical condition index when the pre-warning signaling is triggered is counted, if the number of electrical threshold values is one, the fault shutdown log corresponding to the electrical threshold value is called, the shutdown reason is identified therefrom, and the pre-warning trigger reason of the current pre-warning signaling is taken; If the number of electrical threshold values is greater than one, the electrical threshold values are taken as each group of fuzzy threshold values, the fuzzy threshold value with a higher electrical coefficient between the electrical parameter data in the current time zone is identified by using the analysis algorithm logic, and the fault shutdown log is called, the shutdown reason is identified therefrom, and the pre-warning trigger reason of the current pre-warning signaling is taken.

[0009] Specifically, the specific process of identifying by using the analysis algorithm logic is: The difference between the fuzzy threshold value and the electrical condition index is calculated and the absolute value is taken as the local coefficient; For each group of fuzzy threshold values, the evaluation data packet of the running time zone before the fault occurs is called, and the electrolytic cell voltage and the electrolytic cell current at each time point in the running time zone are divided; The electrolytic cell voltage at each time point in the current time zone is calculated by one-to-one ratio calculation with the electrolytic cell voltage at each time point in the running time zone, the difference between each group of ratios and the integer one is calculated and the absolute value is taken, and the pressure coefficient is obtained by adding up; The electrolytic cell current at each time point in the current time zone is calculated by one-to-one ratio calculation with the electrolytic cell current at each time point in the running time zone, the difference between each group of ratios and the integer one is calculated and the absolute value is taken, and the flow coefficient is obtained by adding up; For the local coefficient, the pressure coefficient and the flow coefficient of each group of fuzzy threshold values, the electrical coefficient is obtained by comprehensive processing using the weighted calculation logic.

[0010] Specifically, the intelligent analysis unit is further used for: comprehensive analysis of the membrane state parameters, determination of the membrane state level of the PEM electrolytic water hydrogen production equipment at the current time point and sending to the cloud monitoring module; wherein the membrane state level includes slight damage, moderate damage and severe damage; the membrane state parameters include membrane humidity, membrane conductivity and membrane thickness; The fluid parameters are analyzed and processed to determine the fluid abnormal state of the PEM electrolytic water hydrogen production equipment at the current time point and sent to the cloud monitoring module; wherein the fluid abnormal state includes a flow rate too high state and a flow rate too low state; the fluid parameter is the water inlet flow rate.

[0011] Specifically, the specific process of determining the membrane state level is: S1: dividing the equipment proton exchange membrane into each monitoring area; S2: comparing the membrane humidity of different monitoring areas with the set humidity threshold, marking the monitoring area below the humidity threshold as a low humidity area; After mean calculation of the membrane humidity of each group of low humidity areas, the humidity threshold is used as the numerator to calculate the ratio, and the humidity deviation value is obtained; the proportion of the number of low humidity areas in the total number of monitoring areas h is calculated to obtain the humidity bias value; S3: similarly to step S2, the membrane thickness is calculated to determine the thickness deviation value and the thickness bias value; Similarly to step S2, the membrane conductivity is calculated to determine the conductivity deviation value and obtain the conductivity bias value; S4: based on the humidity deviation value, the humidity bias value, the thickness deviation value, the thickness bias value, the conductivity deviation value and the conductivity bias value of the equipment proton exchange membrane, the membrane evaluation coefficient is obtained by using the weighted calculation logic for comprehensive processing; the mapping rule between the membrane evaluation coefficient and the membrane state level is constructed in advance, and the mapping conversion is performed to obtain the membrane state level of the PEM electrolytic water hydrogen production equipment at the current time point.

[0012] Specifically, the specific process of determining the fluid abnormal state is: Identify the equipment load state of the PEM electrolytic water hydrogen production equipment at the current time point, including rated load, partial load and low load, and preset the normal flow range corresponding to different equipment load states; Extract the water inlet flow rate of the PEM electrolytic water hydrogen production equipment in the set time zone and perform mean calculation to determine the water inlet average amount; Identify the load state of the equipment in the set time zone, extract the corresponding normal flow range, and compare the water inlet average amount with the normal flow range; if it is higher than the normal flow range, it is determined as a flow rate too high state, otherwise it is determined as a flow rate too low state.

[0013] Specifically, the cloud monitoring module is also used to receive the fluid abnormal state combined with the flow rate change trend for comprehensive processing to determine the current flow rate abnormal reason and send it to the mobile terminal of the operation and maintenance personnel.

[0014] Specifically, the specific process of receiving the fluid abnormal state combined with the flow rate change trend for comprehensive processing to determine the current flow rate abnormal reason is: Establish a historical fault database for excessively high and low flow conditions, and retrieve historical fault data of the same condition from the historical fault database based on the current abnormal fluid condition as reference data. Identify the equipment load status when the reference data fails, and filter data that is the same as the current equipment load status. Plot the influent flow rate curve within the current set time zone, extract the curve of the filtered data, and calculate the root mean square error between the two sets of curves as the curve value. If the flow rate is too high, calculate the difference between the average influent flow rate and the highest value in the normal flow range, and use this as the flow rate difference. If the flow rate is too low, calculate the difference between the average influent flow rate and the lowest value in the normal flow range, and take the absolute value as the flow rate difference. The difference between the flow rate difference of the filtered data and the flow rate difference of the current set time zone is calculated, and the absolute value is taken as the approximate value. For the similarity values ​​and quantitative similarity values ​​of each group of screened data, the data coefficients of each group of screened data are obtained by comprehensive processing using weighted calculation logic. Select the filter data with higher flow coefficients, and retrieve the historical fault causes from them as the causes of current flow anomalies.

[0015] The technical effects and advantages of this invention are as follows: By extracting electrical parameters at the time of fault shutdown and combining them with the operation log to identify the operating condition type at the time of fault, the threshold is ensured to match the operating condition. The electrical parameters before a single fault are divided into evaluation data packages, the voltage deviation rate and current deviation rate are calculated and coordinate points are constructed, and rectangles are drawn with the coordinate points. The area of ​​the rectangle is the electrical state index. Then, electrical threshold sets of low, medium and full load are formed according to the operating condition. The current equipment operating condition and electrical parameters are read in real time. After calculating the current electrical state index, the electrical threshold corresponding to the operating condition is automatically called, which greatly improves the accuracy of early warning and solves the problem that existing technologies mostly rely on fixed rated values ​​to set a single monitoring threshold. By dividing the proton exchange membrane into various monitoring areas, comparing the membrane humidity, membrane thickness, and membrane conductivity of each area, marking low humidity areas, thickness difference areas, and low conductivity areas, and calculating deviation values ​​and bias values, the deviation values ​​and bias values ​​of humidity, thickness, and conductivity are integrated into membrane evaluation coefficients through weighted calculation logic. Then, through preset mapping rules, the damage level is accurately output, realizing a multi-dimensional and quantitative evaluation of membrane status. By determining the current load, calling the normal flow range of the corresponding load, calculating the average inflow of the set time zone, comparing it with the normal range to determine whether the flow is too high or too low, establishing a historical fault database of abnormal flow, first filtering historical data of the same type as the abnormal fluid state and the current load as the filtering data, then weighting and fusing the similarity value and the similarity value to obtain the flow data coefficient, selecting the historical data with the highest flow data coefficient to extract the fault cause as the current abnormal cause, helping maintenance personnel to locate the fault. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] like Figure 1 As shown, a remote safety monitoring system for PEM water electrolysis hydrogen production equipment includes an equipment-side detection module, an edge computing module, and a cloud monitoring module. The equipment monitoring module is used to collect safety data of the PEM water electrolysis hydrogen production equipment in real time using a pre-deployed sensor group. The safety data includes electrical parameters, membrane state parameters and fluid parameters. Electrical parameters include electrolytic cell voltage and electrolytic cell current; Membrane state parameters include membrane humidity, membrane conductivity, and membrane thickness; The fluid parameter is the inlet water flow rate; The sensor array is deployed at the electrolyzer, proton exchange membrane, and inlet pipe of the PEM water electrolysis hydrogen production equipment. 5G / Industrial Ethernet is used to transmit the safety data collected by the sensor group to the data storage unit in real time; The edge computing module includes a data storage unit and an intelligent analysis unit; The data storage unit is used to receive safety data from the PEM water electrolysis hydrogen production equipment and store historical electrical parameter data throughout the entire life cycle of the equipment. The historical electrical parameter data is the electrical parameter data when each group experiences a fault shutdown. The intelligent analysis unit is used to parse safety data, perform comprehensive analysis on electrical parameters and electrical thresholds determined by combining electrical parameters with historical data, and selectively trigger early warning signals to be sent to the cloud monitoring module. Specifically: The electrical parameter data of each group of equipment at the time of failure and shutdown is extracted from the data storage unit, and invalid values ​​caused by sensor malfunctions are removed; the operating condition type of the equipment at the time of failure is identified by the equipment operation log at the time of failure, including low load, medium load and full load. Low load is defined as hydrogen production below 50% of the rated value, medium load as 50% to 80%, and full load as greater than 80%. When a single group of equipment fails and stops, the electrical parameter data of the equipment is divided into electrical parameters of the operating area before the failure by time axis, and used as an evaluation data package; For the evaluation data package, the average values ​​of the electrolytic cell voltage and electrolytic cell current at each time point within the runtime zone are calculated as the total voltage value and the total current value. Calculate the deviation rate between the total voltage and total current and the corresponding rated total voltage and rated total current, and determine the voltage deviation rate and current deviation rate. That is, using formulas (1) and (2) Calculate the voltage deviation rate and current deviation rate ;in and These represent the total voltage and the total current, respectively. and These represent the rated total voltage and rated total current, respectively.

[0019] After representing the voltage deviation rate and current deviation rate as a and b respectively, construct the coordinate point (a, b) and draw a plane rectangular coordinate system. Plot the coordinate point (a, b) in the plane rectangular coordinate system, and draw perpendicular lines from the coordinate point (a, b) to the X-axis and Y-axis respectively until the two perpendicular lines intersect the X-axis and Y-axis respectively. Intercept the figure enclosed by the two perpendicular lines and the X-axis and Y-axis, and calculate the area of ​​the figure as the electrostatic index. In the coordinate system, point P(a,b) is determined. This point is a visual representation of the real-time electrical parameter status. Draw a perpendicular line from point P(a,b) to the X-axis, with the foot of the perpendicular being A(a,0). Draw a perpendicular line from point P(a,b) to the Y-axis, with the foot of the perpendicular being B(0,b). The two perpendicular lines, together with the X-axis and Y-axis, form a rectangle OAPB (O is the origin of the coordinate system).

[0020] After converting and calculating the electrical parameter data of each group that experienced a fault shutdown, the data are classified according to the operating condition of the equipment at the time of each fault, resulting in a low-load index set, a medium-load index set, and a full-load index set. Each set of electrical indices within the low-load index set, medium-load index set, and full-load index set is used as an electrical threshold. The system reads the electrical parameter data and operating condition type of the PEM water electrolysis hydrogen production equipment in the current time zone. After calculating the electrical state index for the electrical parameter data of the PEM water electrolysis hydrogen production equipment in the current time zone, it automatically calls up the electrical thresholds of each group for the corresponding operating condition. The system compares the electrical state index calculated in the current time zone with the electrical thresholds of each group for the corresponding operating condition. If the electrical state index in the current time zone is higher than one of the electrical thresholds, an early warning signal is triggered. Electrical thresholds (sets of electrical indices) are established for low / medium / full load conditions, which solves the misjudgment problem caused by "a single threshold adapting to all operating conditions" in the existing technology; The threshold adaptive electrical threshold based on historical fault data is directly derived from real fault cases under similar operating conditions (electrical index of the operating area before the fault), making the threshold more closely match the actual operating characteristics of the equipment. The threshold will be updated naturally as historical data accumulates.

[0021] A comprehensive analysis of membrane condition parameters is performed to determine the membrane condition level of the PEM water electrolysis hydrogen production equipment at the current time point and send it to the cloud monitoring module; the membrane condition level includes slight damage, moderate damage, and severe damage; Specifically: Divide the proton exchange membrane of the device into h monitoring zones; for example, 5×5=25 zones, and adjust according to the membrane area and sensor accuracy. The membrane humidity and membrane thickness of the proton exchange membrane in different monitoring areas of the equipment were retrieved; The membrane humidity in different monitoring areas is compared with the set humidity threshold, and the monitoring areas below the humidity threshold are marked as low humidity areas. The average humidity of the membrane in each low-humidity area is calculated and used as the denominator. The humidity threshold is used as the numerator to calculate the ratio and obtain the humidity deviation value. The proportion of the number of low-humidity areas in the total number of monitoring areas h is calculated to obtain the humidity deviation value. Using the membrane thickness at the time of manufacture as the baseline thickness, the membrane thickness of different monitoring areas is compared with the baseline thickness, and the monitoring areas with thicknesses lower than the baseline thickness are marked as thickness difference areas; The average thickness of the membrane in each group of thickness difference regions is calculated and used as the denominator. The reference thickness is used as the numerator to calculate the ratio and obtain the thickness deviation value. The proportion of the number of thickness difference regions in the total number of monitoring areas h is calculated to obtain the thickness deviation value. Using the initial conductivity of the new membrane as the reference conductivity, the membrane conductivity of different monitoring areas is compared with the set reference conductivity. Monitoring areas with conductivity lower than the reference conductivity are marked as low-rate areas. The average membrane conductivity of each low-rate region is calculated and used as the denominator. The baseline conductivity is used as the numerator to calculate the ratio, and the conductivity deviation value is obtained. The proportion of the number of low-rate regions in the total number of monitoring regions h is calculated to obtain the conductivity deviation value. Assume that the proton exchange membrane of a PEM water electrolysis hydrogen production device is divided into 5×5=25 monitoring zones (h=25), and the parameters are set as follows: Humidity threshold: 30%; Reference thickness: 175μm (factory thickness); Reference conductivity: 0.08 S / cm (initial value of new membrane); Low humidity areas were identified: the humidity of each area was compared with the 30% threshold, and 9 areas numbered 1-5, 8-10, and 15 were found to have humidity below 30%, which were marked as low humidity areas, with values ​​of 28%, 25%, 26%, 29%, 24%, 27%, 22%, 29%, and 26%, respectively. Humidity deviation = 30% ÷ 25.89% ≈ 1.16; The humidity deviation value = 9 ÷ 25 = 0.36 (i.e., 36%). Marking thickness difference areas: Comparing the thickness of each area with the 175μm benchmark, it was found that the thickness of 9 areas numbered 1-5, 8-10, and 15 was less than 175μm, and these areas were marked as thickness difference areas, with values ​​of 172, 168, 170, 167, 164, 169, 165, 169, and 166, respectively. Thickness deviation = 175 ÷ 167.89 ≈ 1.04; Thickness deviation = 9 ÷ 25 = 0.36 (i.e., 36%); Low-rate regions were identified: the conductivity of each region was compared with the 0.08 S / cm benchmark. Nine regions, numbered 1-5, 8-10, and 15, had conductivity below 0.08 S / cm and were marked as low-rate regions, with values ​​of 0.04, 0.03, 0.035, 0.025, 0.02, 0.03, 0.015, 0.025, and 0.03, respectively. Conductivity deviation = 0.08 ÷ 0.0289 ≈ 2.77; The conductance deflection value = 9 ÷ 25 = 0.36 (i.e. 36%).

[0022] Based on the humidity deviation, humidity deviation, thickness deviation, thickness deviation, conductivity deviation, and conductivity deviation of the proton exchange membrane, a membrane evaluation coefficient is obtained by comprehensive processing using weighted calculation logic. The specific logic for weighted calculation of membrane state coefficients is as follows: The humidity deviation value, humidity deviation value, thickness deviation value, thickness deviation value, conductivity deviation value, and conductivity deviation value are used as follows: The symbol indicates that i represents the number, i=1,2,3,4,5 or6, which correspond to the humidity deviation value, humidity deviation value, thickness deviation value, thickness deviation value, conductivity deviation value and conductivity deviation value, respectively. According to the formula The membrane state coefficient W is obtained after calculation; where Weighting coefficients are set for each group of values; A mapping rule between membrane evaluation coefficients and membrane state levels is pre-constructed, and the mapping transformation is performed to obtain the membrane state level of the PEM water electrolysis hydrogen production equipment at the current time point; That is, the three sets of coefficient value ranges corresponding to the preset membrane state coefficient, each set of coefficient value ranges corresponds to mild damage, moderate damage and severe damage respectively; and the higher the membrane state coefficient, the higher the probability of matching severe damage.

[0023] The fluid parameters are analyzed and processed to determine the abnormal fluid status of the PEM water electrolysis hydrogen production equipment at the current time point and the results are sent to the cloud monitoring module. The abnormal fluid status includes excessively high flow rate and excessively low flow rate. Specifically: Identify the load status of the PEM water electrolysis hydrogen production equipment at the current point in time, including rated load, partial load, and low load, based on the electrolysis current value. For example, an electrolysis current of 200A is the rated load. Preset the normal flow range corresponding to different equipment load statuses. Higher loads require more water to maintain hydration, necessitating a corresponding increase in feed water flow. When the load is stable, the flow rate should be maintained within a narrow range, with preset normal flow rate ranges for different loads, for example: The normal influent flow rate range corresponding to the rated load is 4.5-5.5 L / min; The normal influent flow rate range corresponding to partial load is 2.2-2.8 L / min; Normal influent flow rate range for low load: 0.5-0.8 L / min.

[0024] Extract the influent flow rate of the PEM water electrolysis hydrogen production equipment within a set time zone and calculate the average influent flow rate to determine the average influent flow rate. Identify the load status of the equipment within the set time zone, extract the corresponding normal flow range, compare the average influent flow with the normal flow range, and if it is higher than the normal flow range, it is determined to be a state of excessive flow; otherwise, it is determined to be a state of excessive flow. The cloud monitoring module receives early warning signals and identifies the triggering basis of the early warning signals. It then selectively uses parsing algorithm logic to determine the cause of the early warning and integrates it with the early warning signals to send to the mobile terminal of the operation and maintenance personnel. Specifically: The number of electrical thresholds below the electrical index when the current warning signal is triggered is counted. If the number of electrical thresholds is one, the fault shutdown log corresponding to the electrical threshold is retrieved, and the shutdown cause is identified from it and used as the warning trigger cause for the current warning signal. If the number of electrical thresholds is greater than one, it is used as the fuzzy threshold for each group. The analytical algorithm logic is used to identify the fuzzy threshold with a higher electrical coefficient between it and the electrical parameter data of the current time zone. The fault shutdown log is retrieved, and the shutdown cause is identified from it and used as the warning triggering cause for the current warning signal. For example, if there are two fuzzy thresholds, and the electrical coefficient of fuzzy threshold 1 is less than the electrical coefficient of fuzzy threshold 2, then the fault shutdown log corresponding to fuzzy threshold 2 (assuming the cause is "mild membrane fouling") is retrieved, and the cause of the current warning is determined to be: mild membrane fouling.

[0025] Analysis algorithm logic: The absolute value of the difference between the fuzzy threshold and the electric field index is taken as the local similarity coefficient. For each group of fuzzy thresholds, retrieve the evaluation data packet of the running zone before the fault occurred, and divide the electrolytic cell voltage and electrolytic cell current at each time point in the running zone. The ratio between the electrolytic cell voltage at each time point in the current time zone and the electrolytic cell voltage at each time point in the operating time zone is calculated one by one. The difference between each set of ratios is calculated with an integer, and the absolute value is then summed to obtain the voltage coefficient. For example, the fuzzy threshold compression coefficients: the voltage ratios are 51.2 / 50.5≈1.014, 51.8 / 51≈1.016, 52.3 / 51.5≈1.016, 52.7 / 52≈1.013, and 53.1 / 52.5≈1.011 respectively. The absolute difference between these ratios and 1 is: 0.014, 0.016, 0.016, 0.013, 0.011 → summed = 0.07.

[0026] The ratio between the current of the electrolytic cell at each time point in the current time zone and the current of the electrolytic cell at each time point in the operating time zone is calculated one by one. The difference between each set of ratios is calculated with an integer, and the absolute value is then summed to obtain the flow coefficient. For example, the flow similarity coefficients of the fuzzy threshold 1: the current ratios are 131 / 128≈1.023, 133 / 130≈1.023, 136 / 132≈1.030, 134 / 131≈1.023, 135 / 133≈1.015 respectively. The absolute difference between these values ​​and 1 is: 0.023, 0.023, 0.030, 0.023, 0.015 → summed = 0.114.

[0027] For the local similarity coefficient, compression similarity coefficient, and flow similarity coefficient of each group of fuzzy thresholds, the electrical coefficient is obtained by comprehensive processing using weighted calculation logic. The specific logic for the weighted calculation of electrical coefficients is as follows: The local similarity coefficient, compressive similarity coefficient, and flow similarity coefficient are respectively labeled as... After normalization, according to the formula The membrane state coefficient K was obtained after calculation; These are the weighting coefficients corresponding to the local similarity coefficient, the compressive similarity coefficient, and the flow similarity coefficient.

[0028] It is also used to receive membrane status levels and send them to the mobile terminals of maintenance personnel; It is also used to receive abnormal fluid conditions and combine them with flow rate change trends for comprehensive processing, determine the cause of the current flow rate abnormality, and send it to the mobile terminal of the operation and maintenance personnel. Specifically: Establish a historical fault database for excessively high and low flow conditions, and retrieve historical fault data of the same condition from the historical fault database based on the current abnormal fluid condition as reference data. Historical fault data includes the time of occurrence, cause of fault, and changes in traffic data. Based on the equipment load status when the reference data failed, filter data that is the same as the current equipment load status. Plot the influent flow rate curve within the current set time zone, extract the curve of the filtered data, and calculate the root mean square error between the two sets of curves as the curve value. Specific use of formulas The root mean square error is calculated, where n is the total number of data points. / Let be the value of curve A / B at the i-th time point.

[0029] If the flow rate is too high, calculate the difference between the average influent flow rate and the highest value in the normal flow range, and use this as the flow rate difference. If the flow rate is too low, calculate the difference between the average influent flow rate and the lowest value in the normal flow range, and take the absolute value as the flow rate difference. The difference between the flow rate difference of the filtered data and the flow rate difference of the current set time zone is calculated, and the absolute value is taken as the approximate value. For the similarity values ​​and quantitative similarity values ​​of each group of screened data, the data coefficients of each group of screened data are obtained by comprehensive processing using weighted calculation logic. The specific logic for weighted calculation of data coefficients is as follows: The similarity value and the quantity similarity value are respectively labeled as After normalization, according to the formula The membrane state coefficient N is obtained after calculation; These are the weighting coefficients corresponding to the similarity values ​​and the quantitative similarity values.

[0030] Select the filter data with higher flow coefficients, and retrieve the historical fault causes from them as the causes of current traffic anomalies; For example, searching by "fluid abnormal status": the current status is "excessive flow rate", and three sets of historical faults are filtered out; Filter by "Current Load": The current load is "Rated Load". Filter for loads that are consistent with the fault. Final filtered data: F01 (Control Valve Sticking), F02 (Sensor Drift). The similarity value is quantified by the root mean square error (RMSE) between the "current flow curve" and the "filtered data flow curve". The smaller the RMSE, the more similar the curves are (the better the similarity value). The similarity value is quantified by the absolute difference between the "current traffic difference" and the "filtered data traffic difference". The smaller the difference, the more similar the degree of traffic exceeding the limit (the better the similarity value). Finally, the flow coefficients of the two sets of filtered data are calculated using weighted calculation logic. If the flow coefficient of F01 is higher than that of F02, the historical fault cause of F01 is retrieved, such as the water inlet regulating valve being stuck. The "abnormal state (excessive flow) + abnormal cause (water inlet regulating valve stuck) + handling suggestion (disassemble and clean the regulating valve)" is packaged and sent to the mobile terminal of the operation and maintenance personnel, and simultaneously synchronized to the cloud monitoring module. The above formulas are all dimensionless calculations. Dimensionless calculations can be performed using various methods such as standardization, which will not be elaborated here. The formulas are derived from software simulations based on a large amount of collected data, and the preset parameters in the formulas can be set by those skilled in the art according to the actual situation.

[0031] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, ATA hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state ATA hard disk.

[0032] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0033] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0034] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0035] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0036] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0037] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable ATA hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0038] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A remote safety monitoring system for PEM water electrolysis hydrogen production equipment, characterized in that, Includes the following modules: Equipment-side detection module: Real-time acquisition of safety data from PEM water electrolysis hydrogen production equipment, including electrical parameters, membrane state parameters, and fluid parameters; Edge computing module: includes data storage unit and intelligent analysis unit; Data storage unit: Receives safety data from PEM water electrolysis hydrogen production equipment and stores historical electrical parameter data of the equipment, including historical electrical parameter data of each group when it fails and shuts down; Intelligent analysis unit: After parsing the safety data, it performs a comprehensive analysis of the electrical parameters and the electrical thresholds determined by combining the electrical parameters with historical data, and selectively triggers early warning signals to be sent to the cloud monitoring module; Cloud monitoring module: After receiving the early warning signal and identifying the triggering basis of the early warning signal, it selectively uses the parsing algorithm logic to determine the cause of the early warning and integrates it with the early warning signal to send it to the mobile terminal of the operation and maintenance personnel.

2. The remote safety monitoring system for PEM water electrolysis hydrogen production equipment according to claim 1, characterized in that, The specific process for determining electrical thresholds using historical electrical parameter data is as follows: Electrical parameters include electrolytic cell voltage and electrolytic cell current; Identify the equipment's operating condition type when each set of faults occurs, including low load, medium load, and full load; When a single group of equipment fails and stops, the electrical parameter data of the equipment is divided into electrical parameters of the operating area before the failure by time axis, and used as an evaluation data package; For the evaluation data package, the average values ​​of the electrolytic cell voltage and electrolytic cell current at each time point within the runtime zone are calculated as the total voltage value and the total current value. Calculate the deviation rate between the total voltage and total current and the corresponding rated total voltage and rated total current, and determine the voltage deviation rate and current deviation rate. After representing the voltage deviation rate and current deviation rate as a and b respectively, construct the coordinate point (a, b) and draw a plane rectangular coordinate system. Plot the coordinate point (a, b) in the plane rectangular coordinate system, and draw perpendicular lines from the coordinate point (a, b) to the X-axis and Y-axis respectively until the two perpendicular lines intersect the X-axis and Y-axis respectively. Intercept the figure enclosed by the two perpendicular lines and the X-axis and Y-axis, and calculate the area of ​​the figure as the electrostatic index. After converting and calculating the electrical parameter data of each group that experienced a fault shutdown, the data are classified according to the operating condition of the equipment at the time of each fault, resulting in a low-load index set, a medium-load index set, and a full-load index set. Each set of electrical indices within the low-load index set, medium-load index set, and full-load index set is used as an electrical threshold.

3. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 2, characterized in that, The specific process of selectively triggering the early warning signaling is as follows: The system reads the electrical parameter data and operating condition type of the PEM water electrolysis hydrogen production equipment in the current time zone. After calculating the electrical state index for the electrical parameter data of the PEM water electrolysis hydrogen production equipment in the current time zone, it automatically calls up the electrical thresholds of each group for the corresponding operating condition. The system compares the electrical state index calculated in the current time zone with the electrical thresholds of each group for the corresponding operating condition. If the electrical state index in the current time zone is higher than one of the electrical thresholds, an early warning signal is triggered.

4. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 3, characterized in that, After identifying the triggering criteria for the warning signal, the reason for the warning trigger is selectively determined using the parsing algorithm logic: The number of electrical thresholds below the electrical index when the current warning signal is triggered is counted. If the number of electrical thresholds is one, the fault shutdown log corresponding to the electrical threshold is retrieved, and the shutdown cause is identified from it and used as the warning trigger cause for the current warning signal. If the number of electrical thresholds is greater than one, it is used as the fuzzy threshold for each group. The analytical algorithm logic is used to identify the fuzzy threshold with a higher electrical coefficient between it and the electrical parameter data of the current time zone. The fault shutdown log is retrieved, and the shutdown cause is identified from it and used as the warning triggering cause for the current warning signal.

5. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 4, characterized in that, The specific process of identification using the analytical algorithm logic is as follows: The absolute value of the difference between the fuzzy threshold and the electric field index is taken as the local similarity coefficient. For each group of fuzzy thresholds, retrieve the evaluation data packet of the running zone before the fault occurred, and divide the electrolytic cell voltage and electrolytic cell current at each time point in the running zone. The ratio between the electrolytic cell voltage at each time point in the current time zone and the electrolytic cell voltage at each time point in the operating time zone is calculated one by one. The difference between each set of ratios is calculated with an integer, and the absolute value is then summed to obtain the voltage coefficient. The ratio between the current of the electrolytic cell at each time point in the current time zone and the current of the electrolytic cell at each time point in the operating time zone is calculated one by one. The difference between each set of ratios is calculated with an integer, and the absolute value is then summed to obtain the flow coefficient. For the local similarity coefficient, compression similarity coefficient, and flow similarity coefficient of each group of fuzzy thresholds, the electrical coefficient is obtained by comprehensive processing using weighted calculation logic.

6. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 1, characterized in that, The intelligent analysis unit is also used for: A comprehensive analysis of membrane state parameters is performed to determine the membrane state level of the PEM water electrolysis hydrogen production equipment at the current time point and send it to the cloud monitoring module. The membrane state level includes minor damage, moderate damage, and severe damage. Membrane state parameters include membrane humidity, membrane conductivity, and membrane thickness. The fluid parameters are analyzed and processed to determine the abnormal fluid state of the PEM water electrolysis hydrogen production equipment at the current time point and sent to the cloud monitoring module; The abnormal fluid conditions include excessively high flow rate and excessively low flow rate; the fluid parameter is the inlet flow rate.

7. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 6, characterized in that, The specific process for determining the membrane state level is as follows: S1: Divide the proton exchange membrane of the equipment into various monitoring areas; S2: Compare the membrane humidity of different monitoring areas with the set humidity threshold, and mark the monitoring areas below the humidity threshold as low humidity areas; The average humidity of the membrane in each low-humidity area is calculated and used as the denominator. The humidity threshold is used as the numerator to calculate the ratio and obtain the humidity deviation value. The proportion of the number of low-humidity areas in the total number of monitoring areas h is calculated to obtain the humidity deviation value. S3; Similarly, step S2 is used to calculate the film thickness and determine the thickness deviation value and the thickness deviation value; Similarly, step S2 calculates the membrane conductivity and determines the conductivity deviation value to obtain the conductivity deviation value; S4: Based on the humidity deviation, humidity deviation, thickness deviation, thickness deviation, conductivity deviation, and conductivity deviation of the proton exchange membrane of the equipment, a membrane evaluation coefficient is obtained by comprehensive processing using weighted calculation logic. A mapping rule between membrane evaluation coefficients and membrane state levels is pre-constructed, and the mapping transformation is performed to obtain the membrane state level of the PEM water electrolysis hydrogen production equipment at the current time point.

8. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 6, characterized in that, The specific process for determining the abnormal fluid state is as follows: Identify the equipment load status of the PEM water electrolysis hydrogen production equipment at the current time point, including rated load, partial load and low load, and preset the normal flow range corresponding to different equipment load statuses; Extract the influent flow rate of the PEM water electrolysis hydrogen production equipment within a set time zone and calculate the average influent flow rate to determine the average influent flow rate. The system identifies the load status of the equipment within a set time zone, extracts the corresponding normal flow range, compares the average influent flow with the normal flow range, and determines that the flow is too high if it is higher than the normal flow range, and vice versa.

9. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 6, characterized in that, The cloud monitoring module is also used to receive abnormal fluid status and combine it with flow rate change trends for comprehensive processing, determine the cause of the current flow rate abnormality, and send it to the mobile terminal of the operation and maintenance personnel.

10. A remote safety monitoring system for a PEM water electrolysis hydrogen production equipment according to claim 9, characterized in that, The specific process for determining the cause of the current flow anomaly is as follows: The abnormal state of the received fluid is combined with the flow rate change trend for comprehensive processing. Establish a historical fault database for excessively high and low flow conditions, and retrieve historical fault data of the same condition from the historical fault database based on the current abnormal fluid condition as reference data. Identify the equipment load status when the reference data fails, and filter data that is the same as the current equipment load status. Plot the influent flow rate curve within the current set time zone, extract the curve of the filtered data, and calculate the root mean square error between the two sets of curves as the curve value. If the flow rate is too high, calculate the difference between the average influent flow rate and the highest value in the normal flow range, and use this as the flow rate difference. If the flow rate is too low, calculate the difference between the average influent flow rate and the lowest value in the normal flow range, and take the absolute value as the flow rate difference. The difference between the flow rate difference of the filtered data and the flow rate difference of the current set time zone is calculated, and the absolute value is taken as the approximate value. For the similarity values ​​and quantitative similarity values ​​of each group of screened data, the data coefficients of each group of screened data are obtained by comprehensive processing using weighted calculation logic. Select the filter data with higher flow coefficients, and retrieve the historical fault causes from them as the causes of current flow anomalies.

Images