A diesel generator set controller

Abstract

The application discloses a diesel generator set controller and relates to the technical field of control systems.The controller comprises a multi-core processing unit, a dynamic coupling analysis engine and an instruction arbitrator.The multi-core processing unit is used for efficiently collecting and integrating multi-dimensional real-time data, the dynamic coupling analysis engine is used for constructing an accurate multi-physical field coupling model, and a random forest regression algorithm is used for real-time solving and dynamically correcting weight coefficients, so that high-precision collaborative control under complex working conditions is realized.The instruction arbitrator directly embeds key weight coefficients output by the engine, comprehensively processes priority signals, and intelligently makes decisions and resolves conflicts, so that safety and stability are preferentially ensured in the case of limited resources or target conflicts.A time division multiplexing instruction bus provided with time slot allocation rules provides an efficient, orderly and low-conflict instruction transmission channel, so that key instructions can be reliably sent to an actuator group, and the working condition of the actuator is fed back through a hardware abstraction layer to form a closed loop, thereby improving the safety and reliability of overall operation.

Description

Technical Field

[0001] This invention relates to the field of control system technology, specifically to a diesel generator set controller. Background Technology

[0002] A diesel generator set is an integrated system consisting of a diesel engine, a generator, a control panel, a common base, and other components. As an important power equipment, it plays an irreplaceable role in many fields such as backup power, main power, peak-shaving power, and emergency power supply. Its core function is to convert the mechanical energy of the diesel engine into electrical energy, providing a stable and reliable power supply for various loads. The controller is the "brain" and "nerve center" of the diesel generator set, and its performance directly determines the operating efficiency, reliability, safety, and intelligence level of the entire unit.

[0003] The existing diesel generator set controllers have the following defects:

[0004] 1. Patent document US20120019000A1 discloses a generator set calibration controller; however, the calibration controller in the aforementioned document suffers from the technical problem of being unable to dynamically adjust weight coefficients according to operating conditions and lacking a dynamic coupling analysis engine, resulting in low control accuracy and safety.

[0005] 2. Patent document US08942942B2 discloses a generator set calibration controller, but the calibration controller in the above document has a technical problem of fixed time slot allocation and inability to dynamically allocate resources according to risk coefficient;

[0006] 3. Patent document US6107927A discloses a generator set controller with an overall synchronization mode, but the generator set controller in the above document has technical problems of timing asynchrony and weight update lag.

[0007] 4. Patent document CN102691583A discloses a constant speed and constant frequency dual closed-loop control system and method for a small power diesel generator. However, the device in the above document has the technical problem of lacking hardware coverage channels and blocking of key instructions, resulting in slow fault response. Summary of the Invention

[0008] The purpose of this invention is to provide a diesel generator set controller to solve the technical problems mentioned in the background section.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a diesel generator set controller, comprising a multi-core processing unit, a dynamic coupling analysis engine, and a command arbitrator, wherein the input terminal of the multi-core processing unit is communicatively connected to a sensor group through a hardware abstraction layer, for receiving mechanical dynamic signals, thermodynamic field distribution data, electrical output parameters, and load parameters of the power equipment;

[0010] The data processing terminal of the multi-core processing unit is bidirectionally connected to the dynamic coupling analysis engine for:

[0011] A multi-physics coupling model is constructed based on sensor signals, and the weight coefficients are calculated in real time using the random forest regression algorithm. and Generate multi-actuator collaborative control instructions;

[0012] The weight coefficients of the coupling model are dynamically adjusted based on the system stability feedback value;

[0013] The output of the dynamic coupling analysis engine is directly embedded in the weight decision unit of the instruction arbitrator, outputting... , And according to the preset linear mapping relationship and Convert to thermal stress weighting factor With power stability weighting coefficient ;

[0014] The priority control terminal of the instruction arbitrator is communicatively connected to the output terminals of the fault diagnosis module, the synchronization detection module, and the efficiency optimization module, respectively.

[0015] The output of the instruction arbitrator is communicatively connected to a time-division multiplexed instruction bus.

[0016] The time-division multiplexing instruction bus is configured with time slot allocation rules;

[0017] The time-division multiplexed instruction bus communicates with the control signal input terminal of the actuator group;

[0018] The operating condition feedback terminal of the actuator group is connected to the multi-core processing unit via a hardware abstraction layer.

[0019] Preferably, the power equipment is a diesel generator set, wherein,

[0020] The mechanical dynamic signals include crankshaft speed signals;

[0021] The thermodynamic field distribution data includes cylinder temperature field distribution data;

[0022] The electrical output parameters include the generator output voltage;

[0023] The load parameters include the load current signal.

[0024] Preferably, the time slot allocation rule is as follows:

[0025] The base cycle duration is 5 milliseconds;

[0026] The initial 0-1 millisecond time slot of the cycle is fixedly allocated to the first priority executor;

[0027] The 1-3 millisecond time slots are assigned to the second priority executor by default.

[0028] Periodic 3-5 millisecond time slots are allocated to the third priority actuator;

[0029] When the dynamic coupling analysis engine outputs the thermal stress weighting coefficient With power stability weighting coefficient When the ratio is greater than 2, 50% of the time in that time slot is reallocated to the third priority actuator;

[0030] When the rate of change of the thermodynamic field is greater than 10℃ / s, additional emergency time slots are dynamically allocated to the third priority actuator.

[0031] Preferably, the first priority actuator is a fuel injection valve;

[0032] The second priority actuator is the excitation regulator;

[0033] The third priority actuator is a proportional cooling valve.

[0034] Preferably, the dynamic correction formula for the weighting coefficients is:

[0035]

[0036] in This is the total weight adjustment amount. This represents the system stability deviation. Temperature of the core thermodynamic region For ambient temperature, For the original weighting coefficients of power stability, This is the original weighting coefficient for thermal stress. The rate of change of the stability deviation. It is the reciprocal of the thermodynamic gradient. For differential operators, For time variables, For time differential elements, This is the derivative of the stability deviation.

[0037] Preferably, the system stability deviation is the absolute value of the output power fluctuation (|). |):

[0038] The temperature of the core thermodynamic region is the cylinder temperature.

[0039] When the absolute value of the output power fluctuation is | When | > 0.3%, the dynamic adjustment of the weighting coefficient is triggered.

[0040] Preferably, the fault level output by the fault diagnosis module is a 3-bit binary code, and its most significant bit (MSB) is directly connected to the interrupt request line of the instruction arbiter, and its most significant bit (MSB) is 1; the phase lock flag output by the synchronization detection module is a single-bit signal, which is set when the phase difference is ≤0.3 degrees.

[0041] Preferably, the fault diagnosis module includes a knock analysis unit, the input of which is connected to a cylinder pressure sensor. The knock analysis unit is configured to perform a fast Fourier transform on the cylinder pressure signal.

[0042] When the energy proportion in the 5000Hz to 8000Hz frequency band is ≥40%:

[0043] The hardware overlay channel of the command arbitrator forces the fuel injection valve to reduce the rated fuel supply by 20% within 1 millisecond;

[0044] Simultaneously inject a step increment command into the proportional cooling valve, with the increment value being 30±5% of the current actual cooling flow rate;

[0045] Triggered a non-maskable interrupt signal.

[0046] Preferably, the multi-core processing unit integrates a time-scale alignment module, configured as follows:

[0047] Using mechanical dynamic signals as the reference clock source, hardware-level timestamp synchronization is performed on thermodynamic field distribution data and electrical parameters;

[0048] The synchronized multiphysics data is output to the dynamic coupling analysis engine via a phase-locked loop.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] 1. This invention efficiently collects and integrates multi-dimensional real-time data from mechanical, thermodynamic, electrical, and load dimensions through a multi-core processing unit. It constructs an accurate multi-physics coupling model using a dynamic coupling analysis engine and employs a random forest regression algorithm to solve and dynamically correct weight coefficients in real time. This enables high-precision collaborative control and adaptive optimization under complex operating conditions. The instruction arbitrator directly embeds the key weight coefficients output by the engine and performs intelligent decision-making and conflict arbitration by integrating priority signals from fault diagnosis, synchronization detection, and efficiency optimization modules. This ensures that safety and stability requirements are prioritized when resources are limited or objectives conflict. The time-division multiplexing instruction bus with configured time slot allocation rules provides an efficient, orderly, and low-conflict instruction transmission channel, ensuring that key instructions are reliably delivered to the actuator group. The actuator operating conditions are fed back through the hardware abstraction layer to form a closed loop, driving continuous model optimization and improving the overall safety and reliability of operation.

[0051] 2. This invention ensures a deterministic response in core fuel control by fixing the allocation period from the start of a 0-1 millisecond time slot to the fuel injection valve, based on a thermal stress weighting coefficient. With power stability weighting coefficient When the ratio is greater than 2, 50% of the 1-3 millisecond time slot is dynamically redistributed to the proportional cooling valve to optimize resources to cope with thermal stress risks. When the thermodynamic field change rate is greater than 10℃ / s, an emergency time slot is dynamically generated to enhance the cooling system's emergency response capability to sudden temperature changes. The knock analysis unit accurately diagnoses knock by identifying the energy proportion of the cylinder pressure signal in the 5000-8000Hz frequency band ≥40%, and triggers triple strong real-time linkage. Through the hardware overlay channel, the fuel injection valve is forced to reduce fuel by 20% within 1 millisecond to cut off the knock energy source. Simultaneously, a step incremental command is injected into the proportional cooling valve to quickly enhance cooling. The non-maskable interrupt is triggered to ensure that the command is executed with the highest priority, thereby improving the safety and reliability of the engine under extreme operating conditions.

[0052] 3. This invention achieves multiple optimization effects through the collaborative design of a dynamic weighting coefficient correction formula and a time-scale alignment module. When the absolute value of the output power fluctuation is | When | > 0.3%, the dynamic weighting correction formula is triggered, adjusting the power stability weighting coefficient. With thermal stress weighting coefficient Real-time adaptive adjustment: The time-scale alignment module uses mechanical dynamic signals as the reference clock source. It unifies the time scales of thermodynamic fields and electrical parameters through hardware-level timestamp synchronization and phase-locked loop technology, eliminating timing deviations in multi-physics data. The dynamic coupling analysis engine optimizes weights in real time based on high-confidence synchronized data, triggering time slot reallocation. This significantly improves the control accuracy and real-time response of the system under complex conditions such as load changes and knock risks, fundamentally suppressing power fluctuations and thermal stress risks.

[0053] 4. This invention performs fast Fourier transform spectrum analysis on the cylinder pressure signal through a knock analysis unit and sets a clear energy percentage threshold to achieve precise quantification of knock identification. When severe knock is detected, the hardware overlay channel of the command arbitrator is instantly activated, forcibly reducing the fuel injection valve's fuel supply by 20% within 1 millisecond. Through hardware-level direct intervention, the abnormal combustion energy source is quickly cut off, suppressing knock to the maximum extent. Simultaneously triggered proportional cooling valve step incremental command actively enhances cooling, accelerates the reduction of cylinder high temperature, and collaboratively curbs the deterioration of knock. It also triggers a non-maskable interrupt signal to ensure that this serious fault can be handled immediately with the highest priority for in-depth diagnosis or the initiation of more stringent safety strategies. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the overall system flow of the present invention;

[0055] Figure 2This is a schematic diagram of the time slot allocation process of the present invention;

[0056] Figure 3 This is a schematic diagram of the detonation response process of the present invention;

[0057] Figure 4 This is a schematic diagram of the dynamic coupling analysis engine process of the present invention;

[0058] Figure 5 This is a schematic diagram of the fault diagnosis process of the present invention;

[0059] Figure 6 This is a schematic diagram of the instruction arbitrator process of the present invention;

[0060] Figure 7 This is a schematic diagram of the overall response process of the present invention;

[0061] Figure 8 This is a schematic diagram of the workflow of the present invention.

[0062] In the diagram: 1. Multi-core processing unit; 2. Dynamic coupling analysis engine; 3. Command arbitrator; 4. Hardware abstraction layer; 5. Sensor group; 7. Fault diagnosis module; 8. Synchronous detection module; 9. Efficiency optimization module; 10. Time-division multiplexed command bus; 11. Actuator group; 12. Fuel injection valve; 13. Excitation regulator; 14. Proportional cooling valve; 15. Knock analysis unit; 16. Cylinder pressure sensor; 17. Time scale alignment module. Detailed Implementation

[0063] 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.

[0064] Example 1: Please refer to Figure 1 The present invention provides an embodiment of a diesel generator set controller, comprising a multi-core processing unit 1, a dynamic coupling analysis engine 2, and a command arbitrator 3. The input terminal of the multi-core processing unit 1 is communicatively connected to a sensor group 5 through a hardware abstraction layer 4, for receiving mechanical dynamic signals, thermodynamic field distribution data, electrical output parameters, and load parameters of the power equipment.

[0065] The data processing end of the multi-core processing unit 1 is bidirectionally connected to the dynamic coupling analysis engine 2 for:

[0066] A multi-physics coupling model is constructed based on sensor signals, and the weight coefficients are calculated in real time using the random forest regression algorithm. and Generate multi-actuator collaborative control instructions;

[0067] The weight coefficients of the coupling model are dynamically adjusted based on the system stability feedback value;

[0068] The output of the dynamic coupling analysis engine 2 is directly embedded with the weight decision unit of the instruction arbitrator 3, and the output... , And according to the preset linear mapping relationship and Convert to thermal stress weighting factor With power stability weighting coefficient ;

[0069] The priority control terminal of the instruction arbitrator 3 is communicatively connected to the output terminals of the fault diagnosis module 7, the synchronization detection module 8, and the efficiency optimization module 9, respectively.

[0070] The output of instruction arbiter 3 is connected to time-division multiplexed instruction bus 10.

[0071] The time-division multiplexing instruction bus 10 is configured with time slot allocation rules;

[0072] The time-division multiplexed instruction bus 10 communicates with the control signal input terminal of the actuator group 11;

[0073] The operating condition feedback terminal of actuator group 11 is connected to the multi-core processing unit 1 via hardware abstraction layer 4.

[0074] The power equipment is a diesel generator set, of which,

[0075] Mechanical dynamic signals include crankshaft speed signals;

[0076] Thermodynamic field distribution data includes cylinder temperature field distribution data;

[0077] Electrical output parameters include generator output voltage;

[0078] Load parameters include the load current signal;

[0079] Furthermore, the multi-core processing unit 1 efficiently collects and integrates multi-dimensional real-time data from mechanical, thermodynamic, electrical, and load dimensions. The dynamic coupling analysis engine 2 constructs an accurate multi-physics coupling model, and the random forest regression algorithm is used to calculate and dynamically correct weight coefficients in real time. This achieves high-precision collaborative control and adaptive optimization under complex operating conditions. The instruction arbitrator 3 directly embeds the key weight coefficients output by the engine and integrates the priority signals from the fault diagnosis, synchronization detection, and efficiency optimization modules 9 for intelligent decision-making and conflict arbitration. This ensures that safety and stability requirements are prioritized when resources are limited or objectives conflict. The time-division multiplexing instruction bus 10, configured with time slot allocation rules, provides an efficient, orderly, and low-conflict instruction transmission channel, ensuring that key instructions are reliably delivered to the actuator group 11. The actuator operating conditions are fed back through the hardware abstraction layer 4 to form a closed loop, driving continuous model optimization. This improves the unit's control accuracy, response speed, dynamic adaptability, resource utilization efficiency, and overall operational safety and reliability under complex and variable operating conditions.

[0080] Example 2: Please refer to Figure 2 and Figure 3 One embodiment provided by this invention: the time slot allocation rule is as follows:

[0081] The base cycle duration is 5 milliseconds;

[0082] The initial 0-1 millisecond time slot of the cycle is fixedly allocated to the first priority executor;

[0083] The 1-3 millisecond time slots are assigned to the second priority executor by default.

[0084] Periodic 3-5 millisecond time slots are allocated to the third priority actuator;

[0085] When the thermal stress weighting coefficient output by the dynamic coupling analysis engine 2 With power stability weighting coefficient When the ratio is greater than 2, 50% of the time in that time slot is reallocated to the third priority actuator;

[0086] When the rate of change of the thermodynamic field is greater than 10℃ / s, additional emergency time slots are dynamically allocated to the third priority actuator;

[0087] The first priority actuator is fuel injection valve 12;

[0088] The second priority actuator is the excitation regulator 13;

[0089] The third priority actuator is the proportional cooling valve 14;

[0090] Fault diagnosis module 7 includes a knock analysis unit 15, whose input is connected to cylinder pressure sensor 16. The knock analysis unit 15 is configured to perform a fast Fourier transform on the cylinder pressure signal.

[0091] When the energy proportion in the 5000Hz to 8000Hz frequency band is ≥40%:

[0092] The hardware overlay channel of the command arbitrator 3 forces the fuel injection valve 12 to reduce the rated fuel supply by 20% within 1 millisecond;

[0093] A step increment command is synchronously injected into the proportional cooling valve 14, with the increment value being 30 ± 5% of the current actual cooling flow rate.

[0094] Triggers a non-maskable interrupt signal;

[0095] Furthermore, by fixing the allocation cycle from the start of the 0-1 millisecond time slot to the fuel injection valve 12, a deterministic response of the core fuel control is ensured, based on the thermal stress weighting coefficient. With power stability weighting coefficient When the ratio is greater than 2, 50% of the 1-3 millisecond time slot is dynamically redistributed to the proportional cooling valve 14 to optimize resources and cope with thermal stress risks. When the thermodynamic field change rate is greater than 10℃ / s, an emergency time slot is dynamically generated to enhance the cooling system's emergency response capability to sudden temperature changes. The knock analysis unit 15 accurately diagnoses knock by identifying the energy proportion of the cylinder pressure signal in the 5000-8000Hz frequency band ≥40%, and triggers triple strong real-time linkage. Through the hardware overlay channel, the fuel injection valve 12 is forced to reduce fuel by 20% within 1 millisecond to cut off the knock energy source. Simultaneously, a step increment command is injected into the proportional cooling valve 14 to quickly enhance cooling, and triggers an unmaskable interrupt to ensure that the command is executed with the highest priority, thereby improving the safety and reliability of the engine under extreme operating conditions.

[0096] Example 3: Please refer to Figure 4 and Figure 5 One embodiment provided by this invention: the dynamic correction formula for the weighting coefficients is as follows:

[0097]

[0098] in This is the total weight adjustment amount. This represents the system stability deviation. Temperature of the core thermodynamic region For ambient temperature, For the original weighting coefficients of power stability, This is the original weighting coefficient for thermal stress. The rate of change of the stability deviation. It is the reciprocal of the thermodynamic gradient. For differential operators, For time variables, For time differential elements, The derivative of the stability deviation;

[0099] The system stability deviation is the absolute value of the output power fluctuation (| |):

[0100] The core thermodynamic region temperature is the cylinder temperature;

[0101] When the absolute value of the output power fluctuation is | Dynamic adjustment of the weighting coefficient is triggered when | > 0.3%;

[0102] Multi-core processing unit 1 integrates a time-scale alignment module 17, which is configured as follows:

[0103] Using mechanical dynamic signals as the reference clock source, hardware-level timestamp synchronization is performed on thermodynamic field distribution data and electrical parameters;

[0104] The synchronized multiphysics data is output to the dynamic coupling analysis engine 2 via a phase-locked loop;

[0105] Furthermore, multiple optimization effects are achieved through the collaborative design of the dynamic correction formula for weighting coefficients and the time-scale alignment module 17. When the absolute value of the output power fluctuation is | When | > 0.3%, the dynamic weighting correction formula is triggered, adjusting the power stability weighting coefficient. With thermal stress weighting coefficient Real-time adaptive adjustment: The time-stamp alignment module 17 uses mechanical dynamic signals as the reference clock source. It unifies the time stamps of thermodynamic fields and electrical parameters through hardware-level timestamp synchronization and phase-locked loop technology, eliminating timing deviations in multi-physics data. The dynamic coupling analysis engine 2 optimizes weights in real time based on high-confidence synchronized data, triggering time slot reallocation. This significantly improves the control accuracy and real-time response of the system under complex conditions such as load changes and knock risks, fundamentally suppressing power fluctuations and thermal stress risks.

[0106] Example 4: Please refer to Figure 6 and Figure 7 In one embodiment of the present invention: the fault level output by the fault diagnosis module 7 is a 3-bit binary code, the most significant bit (MSB) of which is directly connected to the interrupt request line of the instruction arbiter 3, and the most significant bit (MSB) of which is 1; the phase lock flag output by the synchronization detection module 8 is a single-bit signal, which is set when the phase difference is ≤0.3 degrees;

[0107] Fault diagnosis module 7 includes a knock analysis unit 15, whose input is connected to cylinder pressure sensor 16. The knock analysis unit 15 is configured to perform a fast Fourier transform on the cylinder pressure signal.

[0108] When the energy proportion in the 5000Hz to 8000Hz frequency band is ≥40%:

[0109] The hardware overlay channel of the command arbitrator 3 forces the fuel injection valve 12 to reduce the rated fuel supply by 20% within 1 millisecond;

[0110] A step increment command is synchronously injected into the proportional cooling valve 14, with the increment value being 30 ± 5% of the current actual cooling flow rate.

[0111] Triggers a non-maskable interrupt signal;

[0112] Furthermore, the cylinder pressure signal is subjected to fast Fourier transform spectrum analysis by the knock analysis unit 15, and a clear energy ratio threshold is set to achieve accurate quantification of knock identification. When severe knock is detected, the hardware overlay channel of the command arbitrator 3 is activated instantly, and the fuel injection valve 12 is forcibly reduced by 20% within 1 millisecond. The abnormal combustion energy source is quickly cut off through hardware-level direct intervention to suppress knock to the maximum extent. The proportional cooling valve 14 is triggered simultaneously to actively enhance cooling with a step incremental command, accelerate the reduction of cylinder temperature, and jointly curb the deterioration of knock. A non-maskable interrupt signal is triggered to ensure that the serious fault can be handled immediately with the highest priority for in-depth diagnosis or the initiation of more stringent safety strategies.

[0113] Example 5: Please refer to Figure 8 The present invention provides an embodiment in which the working steps of the diesel generator set controller are as follows:

[0114] S1. Real-time acquisition of: mechanical dynamic signals, thermodynamic field distribution data, electrical output parameters and load parameters through sensor group 5;

[0115] The time-stamp alignment module 17 uses the crankshaft speed as the reference clock and performs hardware-level timestamp synchronization of multi-source data through a phase-locked loop;

[0116] S2. Construct a multiphysics coupling model based on synchronous data, and use the random forest regression algorithm to calculate the original weight coefficients for power stability in real time. Original weighting coefficient of thermal stress And convert the output control weights according to the formula. and ;

[0117] S3, Output power fluctuation detected | When | > 0.3%, update the weighting coefficients according to the corrected formula:

[0118]

[0119] in Take the actual measured value of the cylinder temperature;

[0120] S4. Allocate instruction time slots every 5 milliseconds:

[0121] The initial 0-1 millisecond time slot of the cycle is fixedly allocated to fuel injection valve 12;

[0122] The 1-3 millisecond time slots are assigned to the excitation regulator 13 by default.

[0123] A 3-5 millisecond time slot is allocated to the proportional cooling valve 14;

[0124] when >2: Transfer 50% of the 1-3ms time period to cooling valve control;

[0125] When the temperature change rate is >10℃ / s: Insert the emergency time slot priority cooling valve for control;

[0126] S5. Knock Detection: Perform fast Fourier transform analysis on the cylinder pressure signal. If the energy in the 5000-8000Hz frequency band is ≥40%, the calibrated fuel supply will be forcibly reduced by 20% within 1ms, and the cooling flow will be increased by 30±5% simultaneously, triggering a non-maskable interruption.

[0127] Fault level coding: The highest bit of the 3-bit binary code is directly connected to the interrupt request line;

[0128] S6, Actuator group 11 operating condition data → fed back to multi-core processing unit 1 in real time through hardware abstraction layer 4, system stability deviation. Participate in the next cycle's weight coefficient adjustment.

[0129] The working principle involves the multi-core processing unit 1 efficiently collecting and integrating multi-dimensional real-time data from mechanical, thermodynamic, electrical, and load dimensions. A precise multi-physics coupling model is constructed using the dynamic coupling analysis engine 2, and a random forest regression algorithm is employed to calculate and dynamically correct weight coefficients in real time. This achieves high-precision collaborative control and adaptive optimization under complex operating conditions. The command arbitrator 3 directly embeds key weight coefficients from the engine output and integrates priority signals from fault diagnosis, synchronization detection, and efficiency optimization modules 9 for intelligent decision-making and conflict arbitration. This ensures that safety and stability requirements are prioritized when resources are limited or objectives conflict. The time-division multiplexing command bus 10, configured with time slot allocation rules, provides an efficient, orderly, and low-conflict command transmission channel, ensuring reliable delivery of key commands to the actuator group 11. Actuator operating conditions are fed back through the hardware abstraction layer 4 to form a closed loop, driving continuous model optimization. This improves the unit's control accuracy, response speed, dynamic adaptability, resource utilization efficiency, and overall operational safety and reliability under complex and variable operating conditions. A fixed allocation period of 0-1 millisecond time slots to the fuel injection valve 12 ensures deterministic response of core fuel control, based on thermal stress weight coefficients. With power stability weighting coefficient When the ratio is greater than 2, 50% of the 1-3 millisecond time slot is dynamically redistributed to the proportional cooling valve 14 to optimize resources and cope with thermal stress risks. When the thermodynamic field change rate is greater than 10℃ / s, an emergency time slot is dynamically generated to enhance the cooling system's emergency response capability to sudden temperature changes. The knock analysis unit 15 accurately diagnoses knock by identifying the energy proportion of the cylinder pressure signal in the 5000-8000Hz frequency band ≥40%, and triggers triple strong real-time linkage. Through the hardware coverage channel, the fuel injection valve 12 is forced to reduce fuel by 20% within 1 millisecond to cut off the knock energy source. Simultaneously, a step increment command is injected into the proportional cooling valve 14 to quickly enhance cooling, and triggers an unmaskable interrupt to ensure the highest priority execution of the command, thereby improving the safety and reliability of the engine under extreme operating conditions. Multiple optimization effects are achieved through the collaborative design of the weighting coefficient dynamic correction formula and the time scale alignment module 17. When the absolute value of the output power fluctuation is | When | > 0.3%, the dynamic weighting correction formula is triggered, adjusting the power stability weighting coefficient. With thermal stress weighting coefficient Real-time adaptive adjustment: The time-stamp alignment module 17 uses the mechanical dynamic signal as the reference clock source. Through hardware-level timestamp synchronization and phase-locked loop technology, it unifies the time stamps of thermodynamic fields and electrical parameters, eliminating timing deviations in multi-physics field data. The dynamic coupling analysis engine 2 optimizes weights in real time based on high-confidence synchronized data, triggering time slot reallocation. This significantly improves the control accuracy and real-time response of the system under complex conditions such as load changes and knock risks, fundamentally suppressing power fluctuations and thermal stress risks. The knock analysis unit 15 performs fast Fourier transform spectrum analysis on the cylinder pressure signal and sets clear parameters. The accurate energy percentage threshold enables precise quantification of knock identification. Upon detecting severe knock, the hardware overlay channel of the command arbitrator 3 is instantly activated, forcibly reducing the fuel supply of the fuel injection valve 12 by 20% within 1 millisecond. Through hardware-level direct intervention, the abnormal combustion energy source is quickly cut off, suppressing knock to the maximum extent. Simultaneously, the proportional cooling valve 14 is triggered to actively enhance cooling with a step incremental command, accelerating the reduction of in-cylinder high temperature, coordinating to curb the deterioration of knock, and triggering a non-maskable interrupt signal to ensure that the severe fault can be handled immediately with the highest priority, for in-depth diagnosis or the initiation of more stringent safety strategies.

[0130] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention.

Claims

1. A diesel generator set controller, comprising a multi-core processing unit (1), a dynamic coupling analysis engine (2), and a command arbitrator (3), characterized in that: The input terminal of the multi-core processing unit (1) is connected to the sensor group (5) through the hardware abstraction layer (4) to receive the mechanical dynamic signals, thermodynamic field distribution data, electrical output parameters and load parameters of the power equipment; The data processing end of the multi-core processing unit (1) is bidirectionally connected to the dynamic coupling analysis engine (2) for: A multi-physics coupling model is constructed based on sensor signals, and the weight coefficients are calculated in real time using the random forest regression algorithm. and Generate multi-actuator collaborative control instructions; The weight coefficients of the coupling model are dynamically adjusted based on the system stability feedback value; The output of the dynamic coupling analysis engine (2) is directly embedded in the weight decision unit of the instruction arbitrator (3), and the output... , And according to the preset linear mapping relationship and Convert to thermal stress weighting factor With power stability weighting coefficient ; The priority control terminal of the instruction arbitrator (3) is respectively connected to the output terminals of the fault diagnosis module (7), the synchronization detection module (8) and the efficiency optimization module (9); The output of the instruction arbiter (3) is connected to the time-division multiplexed instruction bus (10). The time-division multiplexing instruction bus (10) is configured with time slot allocation rules; The time-division multiplexed instruction bus (10) is connected to the control signal input terminal of the actuator group (11); The operating condition feedback terminal of the actuator group (11) is connected to the multi-core processing unit (1) through the hardware abstraction layer (4). The dynamic correction formula for the weighting coefficients is as follows: ； in This is the total weight adjustment amount. This represents the system stability deviation. Temperature of the core thermodynamic region For ambient temperature, For the original weighting coefficients of power stability, This is the original weighting coefficient for thermal stress. The rate of change of the stability deviation. It is the reciprocal of the thermodynamic gradient. For differential operators, For time variables, For time differential elements, This is the derivative of the stability deviation.

2. A diesel generator set controller according to claim 1, characterized in that: The power equipment is a diesel generator set, wherein... The mechanical dynamic signals include crankshaft speed signals; The thermodynamic field distribution data includes cylinder temperature field distribution data; The electrical output parameters include the generator output voltage; The load parameters include the load current signal.

3. A diesel generator set controller according to claim 1, characterized in that: The time slot allocation rule is as follows: The base cycle duration is 5 milliseconds; The initial 0-1 millisecond time slot of the cycle is fixedly allocated to the first priority executor; The 1-3 millisecond time slots are assigned to the second priority executor by default. Periodic 3-5 millisecond time slots are allocated to the third priority actuator; When the dynamic coupling analysis engine (2) outputs the thermal stress weighting coefficient With power stability weighting coefficient When the ratio is greater than 2, 50% of the time in that time slot is reallocated to the third priority actuator; When the rate of change of the thermodynamic field is greater than 10℃ / s, additional emergency time slots are dynamically allocated to the third priority actuator.

4. A diesel generator set controller according to claim 3, characterized in that: The first priority actuator is the fuel injection valve (12); The second priority actuator is the excitation regulator (13); The third priority actuator is the proportional cooling valve (14).

5. A diesel generator set controller according to claim 1, characterized in that: The system stability deviation is the absolute value of the output power fluctuation | |: The temperature of the core thermodynamic region is the cylinder temperature. When the absolute value of the output power fluctuation is | When | > 0.3%, the dynamic adjustment of the weighting coefficient is triggered.

6. A diesel generator set controller according to claim 1, characterized in that: The fault level output by the fault diagnosis module (7) is a 3-bit binary code, and its most significant bit (MSB) is directly connected to the interrupt request line of the instruction arbiter (3). The phase lock flag output by the synchronization detection module (8) is a single-bit signal, which is set when the phase difference is ≤0.3 degrees.

7. A diesel generator set controller according to claim 1, characterized in that: The fault diagnosis module (7) includes a knock analysis unit (15), whose input is connected to a cylinder pressure sensor (16). The knock analysis unit (15) is configured to perform a fast Fourier transform on the cylinder pressure signal. When the energy proportion in the 5000Hz to 8000Hz frequency band is ≥40%: The hardware overlay channel of the command arbitrator (3) forces the fuel injection valve (12) to reduce the calibrated fuel supply by 20% within 1 millisecond; A step increment command is synchronously injected into the proportional cooling valve (14), with the increment value being 30±5% of the current actual cooling flow rate; Triggered a non-maskable interrupt signal.

8. A diesel generator set controller according to claim 1, characterized in that: The multi-core processing unit (1) integrates a time-scale alignment module (17) and is configured as follows: Using mechanical dynamic signals as the reference clock source, hardware-level timestamp synchronization is performed on thermodynamic field distribution data and electrical parameters; The synchronized multiphysics data is output to the dynamic coupling analysis engine via a phase-locked loop (2).

Images