Multi-group water electrolysis hydrogen production system and control method
Optimizing the power regulation of a single-group hydrogen production system through two-layer control system and AI optimization algorithm, the problem of insufficient response capability of large-scale water electrolytic hydrogen production system under renewable energy power fluctuations is solved, and the stability of the system is improved. Cost-effective.
Patent Information
- Application Number
- CN202411254916.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-11-08
AI Technical Summary
When facing the volatility of renewable energy power supply, large-scale water electrolysis hydrogen production systems have insufficient response capabilities, resulting in frequent start and stopping of equipment, increasing costs and poor stability.
Adopt two-layer control system, including parameter setting module, instruction receiving module, perception execution module and multiple control modules, and use AI optimization algorithm and causal forest + machine learning algorithm to optimize the power regulation of a single-group hydrogen production system to reduce the activation The number of stops is improved to improve the system's ability to adapt to power fluctuations.
It improves the adaptability of the water electrolytic hydrogen production system to power fluctuations, reduces the dependence of energy storage batteries, extends the service life of energy storage batteries, and reduces the overall system cost.
Smart Images

Figure CN118910674A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an electrolytic hydrogen production system and a control method, and in particular to a multi-group water electrolytic hydrogen production system and a control method. Background Art
[0002] Hydrogen production by water electrolysis is a method of producing hydrogen by water electrolysis. The basic principle is to pass direct current into an electrolyzer filled with electrolyte, and water molecules undergo electrochemical reactions on the electrodes, decomposing into hydrogen and oxygen. Depending on the electrolyzer diaphragm material, water electrolysis technology can be divided into alkaline water electrolysis (AE), proton exchange membrane (PEM) water electrolysis, and high-temperature solid oxide water electrolysis (SOEC). Currently, alkaline water electrolysis and PEM water electrolysis are widely used and are gradually being promoted.
[0003] A single water electrolysis hydrogen production unit includes a hydrogen production power supply, a water electrolyzer, a separation and purification device, and a water replenishment device. A large-scale water electrolysis hydrogen production system includes multiple water electrolysis hydrogen production units, a water treatment device, a compression control device, etc.
[0004] Renewable energy hydrogen production system will be one of the main application directions of water electrolysis hydrogen production recognized by the national strategy and the market. However, wind / solar power generation is intermittent and uncertain, and hydrogen production equipment needs to adapt to wide power supply fluctuations to a certain extent. The larger the scale of hydrogen production, the more complex the control method.
[0005] To solve this problem, the current solution mainly adopts the start and stop of hydrogen production equipment and the support of energy storage system. On the one hand, the excessive support of energy storage system causes the cost of hydrogen production to rise. On the other hand, because the start and stop time of hydrogen production equipment exceeds the cycle of wide fluctuation of power supply, the coarse-grained start and stop control cannot meet the demand for fluctuation adaptation, which brings stability risks and waste. Summary of the invention
[0006] Purpose of the invention: The purpose of the present invention is to propose a multi-group water electrolysis hydrogen production system and control method to resolve the contradiction between the volatility of renewable energy and the responsiveness of the water electrolysis hydrogen production system, and to enhance the adaptability of the water electrolysis hydrogen production system to power supply fluctuations.
[0007] Technical solution: The hydrogen production system of the present invention adopts a two-layer control system, wherein the first-layer control system includes several single-group control modules, and the second-layer control system includes a parameter setting module, an instruction receiving module, a perception execution module and a plurality of control modules, wherein the output ends of the parameter setting module and the instruction receiving module are connected to the plurality of control modules, the plurality of control modules are interconnected with the perception execution module, and the perception execution module is connected to a plurality of single-group control modules.
[0008] The instruction receiving module receives the scheduling strategy from the superior scheduling system or the production planning system.
[0009] The sensing execution module senses the hydrogen production working state and the operating state of each group of water electrolysis hydrogen production units, and sends control instructions to a single group of control modules.
[0010] The multiple control modules are composed of multiple groups of hydrogen production units to form a unified dispatching execution system. According to the hydrogen production power control value from the upper system, an AI optimization method is used to determine the operating conditions and target power values of each single hydrogen production unit.
[0011] The single-group control module controls and executes a single-group electrolytic cell and a separation and purification device. After receiving the control instruction, it adjusts the single-group power consumption Pa and performs PID adjustment on the corresponding real-time data, with safe production as the highest priority, the lowest oxygen content in hydrogen as the second goal, and the maximum hydrogen production flow rate as the third goal, to achieve single-group control.
[0012] A control method for a multi-group water electrolysis hydrogen production system comprises: obtaining a scheduling target or a preset target from a production schedule or plan; collecting operation data of each unit and device; comparing the deviation between the control instruction and the current system state; calculating the control parameters of each hydrogen production unit / device according to the principle of reducing the start-stop action, and sending them to each single-group control module, integrating the above unit / device operation status list, selecting a controllable hydrogen unit, and calculating the target power of each unit according to the list and the target total power of the hydrogen production power source in the scheduling instruction; the single-group control module controls the equipment and actuator actions of each unit in accordance with the received control parameters.
[0013] The obtaining of the scheduling target or preset target from the production scheduling and planning is specifically as follows: determining whether a periodic scheduling target provided by an external system is received; if a scheduling target provided by an external system is received, setting the calculated hydrogen production power as Pzt; if no scheduling target is received from an external system, according to the preset periodic trend change data of hydrogen production power, selecting the corresponding value at time t+1 as Pzt. .
[0014] The deviation between the comparison control instruction and the current system state specifically includes:
[0015] When the equipment rated power of the equipment state deviation is equal to or greater than a% of the total hydrogen production rated power, the dispatching strategy of the dispatching system will not be executed, and an alarm will be fed back to the superior system; when the equipment rated power of the equipment state deviation is lower than a% of the total hydrogen production rated power, the dispatching strategy of the dispatching system can still be executed;
[0016] The target equipment status is compared with the actual baseline to identify newly added controllable units and shut down units, mark the start and shutdown identifiers of such units, and modify the set IA{I1,A1; I2,A2…} to be used in the next step for power allocation.
[0017] The calculation principle of the target power of each unit is as follows:
[0018] If there are cold standby units or shutdown units, they are preferentially executed;
[0019] Let C i be the allowable adjustable power change amount of each unit within one cycle, and the total controllable power of the hydrogen production system When C ≥ |Pzt – Pzr|, the controllable hydrogen unit can bear the hydrogen production power change in the next cycle, and take Pz t+1 = Pzt; when C < |Pzt – Pzr|, the controllable hydrogen unit cannot bear the hydrogen production power change in the next cycle, and the excess power is fed back to the power control system, energy management system or energy storage control system for consumption or temporary storage, and take Pz t+1 = Pzr ± C;
[0020] According to Pz t+1 , and the current state of each unit and the state change characteristics of the hydrogen production unit, fit the optimal power distribution of each unit;
[0021] According to the calculation model of the change mechanism of the power P and temperature change T of water electrolysis hydrogen production, check whether the temperature change T of each unit exceeds the threshold value. If it exceeds the threshold value, the strategy fails and re-optimization is required; if it does not exceed the threshold value, it means the strategy can be executed.
[0022] The boundary conditions in the fitting of the optimal power distribution of each unit are as follows:
[0023] The total target value in the next cycle is greater than or equal to the sum of the target powers of the devices in set IA, that is:
[0024]
[0025] The absolute value of the power change amount of a single group of hydrogen production units from time t to time t + 1 cannot exceed C of the single group rated power i ;
[0026] The principle of the minimum start-stop times of a single group of hydrogen production units: when Pzr < Pzt, preferentially supplement the power of the operating units, and then consider starting the standby units; when Pzr > Pzt, preferentially reduce the power of the operating units in the same proportion, and then consider shutting down the operating units.
[0027] Beneficial effects: The present invention has the following advantages:
[0028] (1) Based on the power optimization technology of multiple groups of hydrogen production units, under the constraints of multiple objective conditions such as the fluctuation of total power consumption for hydrogen production does not exceed the controllable range, the power fluctuation of a single hydrogen production unit does not exceed the equipment tolerance limit, and the number of unit / equipment start-up and shutdown times is minimized, the causal forest + machine learning algorithm is used to calculate the joint scheduling target of multiple groups of hydrogen production units and the adaptive target of a single group; the control frequency is reasonably increased according to the system scale, and the control scheme of advance adjustment is adopted to improve the power fluctuation adaptability of water electrolysis hydrogen production; in the scenario of renewable energy hydrogen production, the stability of the hydrogen production unit is greatly improved;
[0029] (2) By adopting the technical means of "Internet of Things + AI strategy optimization", the adaptive ability of the hydrogen production unit to power fluctuations is improved through advance analysis and scheduling, thereby realizing the renewable energy allocation and technology of "priority for hydrogen production electricity, and energy storage batteries adjusting the excess", reducing the dependence of renewable energy hydrogen production on energy storage batteries, increasing the service life of energy storage batteries, and reducing the overall cost of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 It is a schematic diagram of the system structure of the present invention;
[0031] Figure 2 It is a flow chart of the present invention. DETAILED DESCRIPTION
[0032] The present invention will be further described below in conjunction with the accompanying drawings.
[0033] like Figure 1 As shown, the multi-group water electrolysis hydrogen production system of the present invention adopts a two-layer control system, a first-layer control system includes several single-group control modules, and a second-layer control system includes a parameter setting module, an instruction receiving module, a perception execution module and a multi-group control module, wherein the output ends of the parameter setting module and the instruction receiving module are connected to the multi-group control module, the multi-group control module is interconnected with the perception execution module, and the perception execution module is connected to multiple single-group control modules through a bus. A set of devices is provided with N groups of centralized water electrolysis hydrogen production units, and the device carrier can be a distributed controller (DCS), a monitoring configuration software (SCADA), a centralized controller, an embedded AI edge intelligent control machine, an Internet of Things configuration system, etc. Each group of water electrolysis hydrogen production units is equipped with a set of single-group control modules, and the module carrier can be a programmable controller (PLC), a programmable edge computing device, an embedded industrial computer, a remote terminal unit (DTU), etc.
[0034] The parameter setting module is used to set the characteristics of the hydrogen production system, equipment parameters, IO points, calculation models, etc.; the characteristics of the hydrogen production system are set, including clocks, uplink / downlink cycles, upper and lower limits of hydrogen production power, fault settings, equipment control point tables, command receiving message formats, etc.; the calculation model configuration is based on the actual scale and equipment conditions, and a single hydrogen production unit, water treatment device, air compression system, etc. are instantiated and configured. The equipment parameters configure the hydrogen production power supply, a single water electrolysis hydrogen production unit, metering meters (pressure, temperature, flow, dew point, etc.), water treatment device, air compression device, etc., and configure the operating parameters and nameplate parameters respectively. According to the calculation model instance, IO points are equipped to correspond to the input and output data points of a single control module, superior scheduling or production planning system.
[0035] The command receiving module receives the scheduling strategy from the superior scheduling system or production planning system, including power supply distribution data, controllable and shut-down equipment lists; the command receiving module has functions such as security authentication, message parsing, message sending, uplink interface management, and communication log. It transmits data with the superior system according to the uplink cycle configured by parameters, receives commands, and reports operating data. Security authentication encrypts and decrypts the communication request and message header of the superior system to ensure the security of IoT communication; message parsing parses the received message into the synchronous control data format; sending messages completes the reporting of system and equipment operating status and operating data, and feedbacks the message of whether the command is received successfully or not; interface management manages the IP address, port, data transmission mode, master-slave mode, etc. of the opposite system; communication log records the data transmission record with the opposite system.
[0036] The sensing execution module senses the hydrogen production status and the operating status of each group of water electrolysis hydrogen production units, and sends the control instructions to the single group control module. The sensing execution module includes functions such as downlink interface management, data acquisition, communication alarm, and parameter delivery. Downlink interface management manages the communication address, data acquisition frequency, IO binding point relationship and other parameters between the single group control module and other equipment controllers; data acquisition, according to the configured data acquisition frequency, obtains operating data from the lower-level controller in sequence; communication alarm, monitors communication interruption, data loss, data limit exceeding and other situations and sends alarm signals; parameter delivery, sends the control parameters calculated by multiple control modules to the specified controller.
[0037] Multiple groups of control modules and multiple groups of hydrogen production units constitute a unified scheduling execution system. Based on the hydrogen production power control value from the superior system (production planning, energy management, power scheduling, virtual power plant and other systems), the AI optimization method is used to determine the operating conditions and target power values of each individual hydrogen production unit, thereby realizing the scheduling control of multiple groups of hydrogen production units.
[0038] The single group control module controls and executes the single group electrolyzer and separation and purification device. After receiving the control command, it slowly adjusts the single group power consumption Pa, controls the hydrogen pressure, outlet liquid temperature, inlet and outlet temperature difference, oxygen content in hydrogen and other real-time data for PID adjustment, and ensures dynamic production safety and balance. With safe production as the highest priority, the lowest oxygen content in hydrogen as the second goal, and the maximum hydrogen production flow as the third goal to achieve single group control.
[0039] like Figure 2 As shown, the control method of the multi-group water electrolysis hydrogen production system of the present invention comprises the following steps:
[0040] S1 obtains the scheduling target (strategy of the superior system) or preset target from production scheduling and planning;
[0041] The scheduling objectives include the list of controllable equipment - set IA (including equipment parameters, equipment status, whether it is faulty, etc.), the total target value Qzt of hydrogen production electricity consumption within the control period, the target value Pzt of hydrogen production power, etc.
[0042] S11 determines whether a periodic dispatch target is received from an external system. The dispatch target can be the hydrogen production power allocated by the power control system, or the hydrogen production power allocated by the plant-wide energy management system, or the hydrogen production power calculated from the hydrogen output of the production plan and the power generation / consumption forecast data;
[0043] S12: if a dispatch target is received from an external system, the calculated hydrogen production power is set as Pzt;
[0044] If S13 does not receive the dispatch target provided by the external system, the value at the corresponding time t+1 is selected as Pzt according to the preset periodic trend change data of the hydrogen production power.
[0045] S2 collects the operating data of each unit and equipment;
[0046] The operation data of each unit and equipment include the equipment operation status (operation, hot standby, cold standby, shutdown), the current power consumption, pressure, temperature, flow rate, etc. of each unit. The real-time total power of hydrogen production and the real baseline status of the equipment (mainly based on temperature parameters) are also counted.
[0047] The unit refers to the hydrogen production unit, and the equipment includes public auxiliary equipment such as air compressor, circulating water, and make-up water.
[0048] Real-time total power of hydrogen production Where Pa represents the actual power consumption of the hydrogen production unit; P b Indicates the power consumption of other public auxiliary equipment.
[0049] The real baseline status of the equipment refers to the current working conditions of each unit (cold standby / hot standby, operation, fault shutdown), and the operating status of the water electrolyzer equipment in the unit (cold standby / hot standby, operation, fault shutdown). The state set of the unit and equipment is set to A{cold standby, hot standby, operation, fault shutdown}.
[0050] S3 compares the deviation between the control instruction and the current system state;
[0051] S31. Baseline equipment state deviation (specifically water electrolyzer equipment) refers to whether the baseline state in the dispatch instruction deviates from the actual baseline state. If there is a deviation, it means that the dispatch system has identified an error and reports the real-time status to the superior system. When the equipment rated power of the equipment state deviation is equal to or greater than a% of the total hydrogen production rated power, the dispatch strategy of the dispatch system will not be executed, and an alarm will be fed back to the superior system. When the equipment power of the equipment state deviation is lower than a% of the total hydrogen production rated power, the dispatch strategy of the dispatch system can still be executed (the deviation equipment is brought to the next step according to the real baseline). a% is based on the comprehensive assessment of the controllable total power of the overall hydrogen production system and the ability of the water electrolyzer to adjust the power. In principle, when a% is evenly distributed to N-1 electrolyzers, it will not cause a collapse in system stability.
[0052] S32. Compare the target equipment status with the real baseline, identify the newly added controllable units and shut-down units, and mark the start-up and shutdown marks of such units. The modified set IA{I1,A1;I2,A2…} is brought into the next step for power allocation.
[0053] S4 calculates the control parameters of each hydrogen production unit / equipment according to the principle of reducing start-stop actions, and sends them to each single group control module; integrates the above unit / equipment operation status list, selects the controllable hydrogen unit, and calculates the target power of each unit according to the list and the total target power of hydrogen production power in the dispatch instruction according to the following principles:
[0054] S41 will be executed first if there is a cold standby unit or parking unit.
[0055] S42 Set C i The total controllable power of the hydrogen production system is the power variation allowed for each unit in one cycle.
[0056]
[0057] When C>=|Pzt–Pzr|, the controllable hydrogen unit can bear the change of hydrogen production power in the next cycle, and Pz t+1 =Pzt;
[0058] When C<|Pzt–Pzr|, the controllable hydrogen unit cannot bear the change of hydrogen production power in the next cycle, and the excess power needs to be fed back to the power control system, energy management system or energy storage control system for consumption or temporary storage.t+1 = Pzr ± C; (Pzr < Pzt, use addition), (Pzr > Pzt, use subtraction)
[0059] S43 is based on Pz t+1 , and the current status of each unit and the characteristics of the change in the status of the hydrogen production unit, to fit the optimal power distribution of each unit.
[0060] Boundary conditions:
[0061] ① The total target value for the next cycle should be slightly greater than or equal to the sum of the target powers of the devices in set IA:
[0062]
[0063] ② The absolute value of the power change of a single hydrogen production unit from time t to time t + 1 should not exceed C of the rated power of a single unit i , C i is determined according to the wide-load fluctuation ability of the water electrolysis cell;
[0064] ③ The principle of minimizing the start-stop times of a single hydrogen production unit: When Pzr < Pzt, give priority to supplementing the power of the operating units, and then consider starting the standby units; when Pzr > Pzt, give priority to proportionally reducing the power of the operating units, and then consider stopping the operating units (the order of stopping, the units with higher cumulative working hours are given priority to stop).
[0065] The input parameters include: set IA, Pzr, Pz t+1 , the real-time operating parameters of the hydrogen production unit, and the model configuration information.
[0066] Basic algorithm: Causal Forest + Machine Learning algorithm. The Causal Forest algorithm is used to set parameter optimization according to empirical data and physical mechanisms for analysis, and the Machine Learning algorithm is used to optimize after training with historical data samples and is optimized as the on-site data accumulates continuously.
[0067] Optimization calculation result:
[0068] Optimal result (minimum start-stop of units): The array of power target values of each unit {Pa 1,t+1 , Pa 2,t+1 , … Pa n,t+1}
[0069] S44 checks whether the temperature change T of each unit exceeds the threshold according to the calculation model of the change mechanism of the power P and temperature change T of water electrolysis hydrogen production. If it exceeds the threshold, the strategy fails and re-optimization is required; if it does not exceed the threshold, it means the strategy can be executed.
[0070] Send the power target values of each unit to each single-group control module according to the set cycle and downlink channel.
[0071] The S5 single control module follows the received control parameters to control the equipment and actuators of each unit. The efficiency of the single hydrogen production unit is controlled at 30% to 110% (adjusted and set according to the test data of the water electrolyzer). After receiving the control parameters from multiple control modules, the single control module controls the hydrogen production power supply, separation and purification device, valve, booster pump, etc. through the control program and judgment logic, and uses the PID regulator to control the single water electrolysis hydrogen production unit to smoothly achieve the control parameter requirements.
Claims
1. A multi-group water electrolysis hydrogen production system, It is characterized in that The hydrogen production system adopts a two-layer control system. The first-layer control system includes several single-group control modules, and the second-layer control system includes a parameter setting module, an instruction receiving module, a perception execution module and a multi-group control module. The output ends of the parameter setting module and the instruction receiving module are connected to the multi-group control modules, the multi-group control modules are interconnected with the perception execution module, and the perception execution module is connected to multiple single-group control modules.
2. A multi-group water electrolysis hydrogen production system according to claim 1, It is characterized in that The instruction receiving module receives the scheduling strategy from the superior scheduling system or the production planning system.
3. A multi-group water electrolysis hydrogen production system according to claim 1, It is characterized in that The sensing execution module senses the hydrogen production working state and the operating state of each group of water electrolysis hydrogen production units, and sends control instructions to a single group of control modules.
4. A multi-group water electrolysis hydrogen production system according to claim 3, It is characterized in that The multiple control modules are composed of multiple groups of hydrogen production units to form a unified dispatching execution system. According to the hydrogen production power control value from the upper system, an AI optimization method is used to determine the operating conditions and target power values of each single hydrogen production unit.
5. A multi-group water electrolysis hydrogen production system according to claim 3, It is characterized in that After receiving the control instruction, the single-group control module adjusts the single-group power consumption Pa and performs PID adjustment on the corresponding real-time data, with safe production as the highest priority, the lowest oxygen content in hydrogen as the second goal, and the maximum hydrogen production flow rate as the third goal, to achieve single-group control.
6. A control method for a multi-group water electrolysis hydrogen production system, It is characterized in that include: Obtain scheduling targets or preset targets from production scheduling and planning; Collect operation data of each unit and equipment; Compare the deviation between control instructions and current system status; According to the principle of reducing start-stop actions, the control parameters of each hydrogen production unit / equipment are calculated and sent to each single group control module. The above unit / equipment operation status list is integrated, and the controllable hydrogen unit is selected. According to the list and the total target power of hydrogen production power source in the dispatching instruction, the target power of each unit is calculated; A single group of control modules follows the received control parameters to control the equipment and actuator actions of each unit.
7. A control method for a multi-group water electrolysis hydrogen production system according to claim 6, It is characterized in that The obtaining of the scheduling target or preset target from the production scheduling and plan is specifically as follows: determining whether a periodic scheduling target provided by an external system is received; if a scheduling target provided by an external system is received, setting the calculated hydrogen production power as Pzt; if no scheduling target is received from the external system, according to the preset periodic trend change data of the hydrogen production power, selecting the value at the corresponding time t+1 as Pzt.
8. A control method for a multi-group water electrolysis hydrogen production system according to claim 6, It is characterized in that The deviation between the comparison control instruction and the current system state specifically includes: When the rated power of the device with a device status deviation is equal to or greater than a% of the total hydrogen production rated power, the scheduling strategy of the scheduling system is not executed, and an alarm is sent to feedback to the superior system; when the rated power of the device with a device status deviation is lower than a% of the total hydrogen production rated power, the scheduling strategy of the scheduling system can still be executed; Compare the target device status with the true baseline, identify the newly added controllable units and shutdown units, mark the start and stop identifiers of such units, and revise the set IA{I1,A1;I2,A2…} and bring it into the next step for power distribution.
9. A control method for a multi-group water electrolysis hydrogen production system according to claim 7, characterized in that the calculation principle of the target power of each unit is as follows: If there are cold standby units or shutdown units, they are given priority; Let C i The total controllable power of the hydrogen production system is the power variation allowed for each unit in one cycle. When C ≥ |Pzt–Pzr|, the controllable hydrogen unit can bear the change of hydrogen production power in the next cycle, and Pz t+1 =Pzt; when C<|Pzt–Pzr|, the controllable hydrogen unit cannot bear the change of hydrogen production power in the next cycle, and the excess power is fed back to the power control system, energy management system or energy storage control system for consumption or temporary storage, and Pz is taken. t+1 =Pzr±C; According to Pz t+1 , as well as the current state of each unit and the state change characteristics of the hydrogen production unit, to fit the optimal power allocation of each unit; According to the calculation model of the change mechanism of the power P and temperature change T of water electrolysis hydrogen production, check whether the temperature change T of each unit exceeds the threshold. If it exceeds the threshold, the strategy fails and re-optimization is required; if it does not exceed the threshold, it means the strategy can be executed.
10. A control method for a multi-group water electrolysis hydrogen production system according to claim 9, characterized in that the boundary conditions in the optimal power distribution fitting of each unit are as follows: The total target value of the next cycle is greater than or equal to the sum of the target powers of the devices in set IA, that is: The absolute value of the power change of a single hydrogen production unit from time t to time t+1 cannot exceed the C of the rated power of the single unit. i ; The principle of minimizing the start-stop times of a single hydrogen production unit: when Pzr < Pzt, give priority to supplementing the power of the operating units, and then consider starting the standby units; when Pzr > Pzt, give priority to proportionally reducing the power of the operating units, and then consider shutting down the operating units.
Citation Information
Cited By
Alkaline electrolytic water hydrogen production energy management system based on superposition and coupling of multiple algorithms, hydrogen production dynamic optimization method and electronic equipment
CN121596723A
